mkdir -p /home/daytona/workspace/cxr-handbook && ls /home/daytona/workspace/

mkdir -p /home/daytona/workspace/cxr-handbook && python3 -c "import reportlab; print('reportlab ok')" && python3 -c "import requests; print('requests ok')"

~/cxr-handbook/build_handbook.py

mkdir -p /home/daytona/workspace/cxr-handbook && python3 -c "import reportlab, PIL, requests; print('All libs OK')"

cat > /home/daytona/workspace/cxr-handbook/build.py << 'PYEOF' import requests, io, os from reportlab.lib.pagesizes import A4 from reportlab.lib import colors from reportlab.lib.units import cm from reportlab.lib.styles import ParagraphStyle from reportlab.lib.enums import TA_CENTER, TA_LEFT, TA_JUSTIFY from reportlab.platypus import (SimpleDocTemplate, Paragraph, Spacer, Image, Table, TableStyle, PageBreak, HRFlowable) from PIL import Image as PILImage OUT = "/home/daytona/workspace/cxr-handbook/CXR_Handbook_Intern.pdf" W, H = A4 # Colors DARK_BLUE = colors.HexColor("#1A3A5C") MED_BLUE = colors.HexColor("#2E6DA4") LIGHT_BLUE = colors.HexColor("#D6E8F7") TEAL = colors.HexColor("#1B7A6E") LIGHT_TEAL = colors.HexColor("#D0EFEB") RED = colors.HexColor("#C0392B") LIGHT_RED = colors.HexColor("#FADBD8") ORANGE = colors.HexColor("#E67E22") LIGHT_ORANGE= colors.HexColor("#FDEBD0") PURPLE = colors.HexColor("#6C3483") LIGHT_PURPLE= colors.HexColor("#E8DAEF") GREEN = colors.HexColor("#1E8449") LIGHT_GREEN = colors.HexColor("#D5F5E3") GOLD = colors.HexColor("#D4AC0D") LIGHT_GOLD = colors.HexColor("#FEF9E7") DARK_GREY = colors.HexColor("#2C3E50") MID_GREY = colors.HexColor("#7F8C8D") LIGHT_GREY = colors.HexColor("#F2F3F4") def S(name, **kw): return ParagraphStyle(name, **kw) sb = S("b", fontName="Helvetica", fontSize=9.5, leading=14, textColor=DARK_GREY, alignment=TA_JUSTIFY, spaceAfter=5) sc = S("c", fontName="Helvetica-Oblique",fontSize=8, leading=11, textColor=MID_GREY, alignment=TA_CENTER, spaceAfter=4) sbul = S("u", fontName="Helvetica", fontSize=9, leading=13, textColor=DARK_GREY, leftIndent=14, spaceAfter=3) sk = S("k", fontName="Helvetica-Bold", fontSize=9, leading=13, textColor=DARK_BLUE, leftIndent=14, spaceAfter=3) def p(t, s=None): return Paragraph(t, s or sb) def bu(t): return Paragraph(f"• {t}", sbul) def kp(t): return Paragraph(f"➤ {t}", sk) def SP(n=0.3): return Spacer(1, n*cm) def HR(c=LIGHT_BLUE): return HRFlowable(width="100%", thickness=2, color=c, spaceAfter=5, spaceBefore=4) def fetch(url, w=10*cm, h=7*cm): try: r = requests.get(url, timeout=20) if r.status_code != 200: return None buf = io.BytesIO(r.content) pil = PILImage.open(buf); iw,ih = pil.size asp = ih/iw if asp*w > h: w = h/asp else: h = asp*w buf.seek(0) img = Image(buf, width=w, height=h); img.hAlign='CENTER' return img except: return None def hdr(title, sub="", col=MED_BLUE, num=""): bg = col rows = [[p(f"<b><font color='white'> {num} {title}</font></b>" if num else f"<b><font color='white'> {title}</font></b>", S("hh", fontName="Helvetica-Bold", fontSize=13, leading=16, textColor=colors.white))]] if sub: rows.append([p(f"<font color='#CCEEFF' size='9'><i> {sub}</i></font>", S("hs", fontName="Helvetica-Oblique", fontSize=9, textColor=colors.HexColor("#CCEEFF")))]) t = Table(rows, colWidths=[W-4*cm]) t.setStyle(TableStyle([ ('BACKGROUND',(0,0),(-1,-1),bg), ('TOPPADDING',(0,0),(-1,-1),8),('BOTTOMPADDING',(0,0),(-1,-1),8), ('LEFTPADDING',(0,0),(-1,-1),12),('RIGHTPADDING',(0,0),(-1,-1),12), ])) return t def subhdr(txt, col=MED_BLUE): t = Table([[p(f"<b><font color='white'>{txt}</font></b>", S("sh", fontName="Helvetica-Bold", fontSize=10, textColor=colors.white))]], colWidths=[W-4*cm]) t.setStyle(TableStyle([ ('BACKGROUND',(0,0),(-1,-1),col), ('TOPPADDING',(0,0),(-1,-1),5),('BOTTOMPADDING',(0,0),(-1,-1),5), ('LEFTPADDING',(0,0),(-1,-1),10), ])) return t def box(items, bg=LIGHT_BLUE, border=MED_BLUE): rows = [[i] for i in items] t = Table(rows, colWidths=[W-5*cm]) t.setStyle(TableStyle([ ('BACKGROUND',(0,0),(-1,-1),bg), ('BOX',(0,0),(-1,-1),1.5,border), ('TOPPADDING',(0,0),(-1,-1),5),('BOTTOMPADDING',(0,0),(-1,-1),5), ('LEFTPADDING',(0,0),(-1,-1),10),('RIGHTPADDING',(0,0),(-1,-1),10), ])) return t def two_col(l_items, r_items, lh="CXR Signs", rh="Clinical Notes", hcol=MED_BLUE): cw = (W-5*cm)/2 data = [[p(f"<b><font color='white'>{lh}</font></b>",sb), p(f"<b><font color='white'>{rh}</font></b>",sb)]] for i in range(max(len(l_items),len(r_items))): l = l_items[i] if i<len(l_items) else "" r = r_items[i] if i<len(r_items) else "" data.append([bu(l), bu(r)]) t = Table(data, colWidths=[cw,cw]) t.setStyle(TableStyle([ ('BACKGROUND',(0,0),(-1,0),hcol), ('ROWBACKGROUNDS',(0,1),(-1,-1),[colors.white,LIGHT_GREY]), ('BOX',(0,0),(-1,-1),1,hcol), ('INNERGRID',(0,0),(-1,-1),0.5,colors.HexColor("#CCCCCC")), ('TOPPADDING',(0,0),(-1,-1),5),('BOTTOMPADDING',(0,0),(-1,-1),5), ('LEFTPADDING',(0,0),(-1,-1),7),('RIGHTPADDING',(0,0),(-1,-1),7), ])) return t def grid(headers, rows_data, col_widths, hcol=DARK_BLUE, alt=LIGHT_BLUE): data = [[p(f"<b><font color='white'>{h}</font></b>",sb) for h in headers]] for row in rows_data: data.append([p(str(c),sb) for c in row]) t = Table(data, colWidths=col_widths) t.setStyle(TableStyle([ ('BACKGROUND',(0,0),(-1,0),hcol), ('ROWBACKGROUNDS',(0,1),(-1,-1),[colors.white,alt]), ('BOX',(0,0),(-1,-1),1,hcol), ('INNERGRID',(0,0),(-1,-1),0.5,colors.HexColor("#BBBBBB")), ('TOPPADDING',(0,0),(-1,-1),5),('BOTTOMPADDING',(0,0),(-1,-1),5), ('LEFTPADDING',(0,0),(-1,-1),7),('RIGHTPADDING',(0,0),(-1,-1),7), ('VALIGN',(0,0),(-1,-1),'MIDDLE'),('FONTSIZE',(0,0),(-1,-1),8.5), ])) return t # Image URLs from Murray & Nadel, Grainger & Allison, Goldman-Cecil IMGS = { "lobar_pna": "https://cdn.orris.care/cdss_images/4107a235bcb72d12fd7e03be9f1a5dafdd7494a8813a4916ae69a9055048eebd.png", "atypical_pna": "https://cdn.orris.care/cdss_images/b5c9f16401022bfb2d51e0fdd2637239387d6efed5968b9937a273228a46936e.png", "aspiration": "https://cdn.orris.care/cdss_images/f97f914e6eb5f10ade11a5d80ffbc4d3088ed5b668ac75125607e5420e132c04.png", "klebsiella": "https://cdn.orris.care/cdss_images/3ea63f4fb73b96ca26369c7316999576edae82a8414da54f704a7640fa9a8432.png", "abscess": "https://cdn.orris.care/cdss_images/d45aba146951a6d8da1d452da56f5b3ddaf43e36994cf7e7f72f04d0d1bc197f.png", "covid_early": "https://cdn.orris.care/cdss_images/9262c9349b671f32e5a28ad645bff170da0f711b5cec9534372b648201998e69.png", "covid_late": "https://cdn.orris.care/cdss_images/6825410929b7c2ca0df925a29c2fede605e83e962d14298226fc12afa31c1215.png", "tb_primary": "https://cdn.orris.care/cdss_images/a8d4518bc7b3478b3ecea071bce32212f29a89347dd7529573da89d558556a8e.png", "tb_cavitary": "https://cdn.orris.care/cdss_images/77ba4b612dfbb87572a123ad3217ba95e9e97f4c32f08a9d5586b4020604a0d3.png", "tb_miliary": "https://cdn.orris.care/cdss_images/ddac82cec1d3bd545a6e03713c64e9cb5bee36b65f75ee822aa6f0d506c90e97.png", "ptx_insp": "https://cdn.orris.care/cdss_images/3848a612ca4563d4d4b97a4f0ff7d757f1385dcf0fcf6d2d0daf9775e39f3d99.png", "ptx_exp": "https://cdn.orris.care/cdss_images/85971adc7eb8b4ea0487b6b1b899f501cc0ebc7ef8022e454e23c462b74d33dc.png", "ptx_skinfold": "https://cdn.orris.care/cdss_images/e088349db48b2c6f428ba8a29aa3031460df630929623c6504da45a4750c48ac.png", "effusion": "https://cdn.orris.care/cdss_images/c67494bd5d0ea8fa634b60d51c5efad65d8de9bdaf7367f31794b662fc55f0ca.png", "edema": "https://cdn.orris.care/cdss_images/feeb5dadaec1ba52bb94d97bbc9164a393847675bc322c0e53e5c4adbe4b62a9.png", } print("Fetching images...") imgs = {} for k,u in IMGS.items(): imgs[k] = fetch(u, 10*cm, 7*cm) print(f" {'OK' if imgs[k] else 'FAIL'} {k}") def img_cap(key, cap, w=10*cm, h=7*cm): out = [] if imgs.get(key): i2 = fetch(IMGS[key], w, h) if i2: out.append(i2) out.append(p(cap, sc)) return out # ── COVER via onFirstPage ───────────────────────────────────────────────────── def cover_page(c, doc): c.saveState() c.setFillColor(DARK_BLUE); c.rect(0,0,W,H,fill=1,stroke=0) c.setFillColor(MED_BLUE); c.rect(0,H*0.52,W,H*0.48,fill=1,stroke=0) c.setFillColor(TEAL); c.rect(0,H*0.49,W,0.04*H,fill=1,stroke=0) c.setFillColor(GOLD); c.rect(0,H*0.46,W,0.03*H,fill=1,stroke=0) c.setFillColor(colors.white); c.setFont("Helvetica-Bold",38) c.drawCentredString(W/2, H*0.65, "CHEST X-RAY") c.setFont("Helvetica-Bold",28) c.drawCentredString(W/2, H*0.59, "CLINICAL HANDBOOK") c.setFillColor(GOLD); c.setFont("Helvetica-Bold",13) c.drawCentredString(W/2, H*0.545, "Pulmonary Department — Intern Physician Edition") c.setFillColor(colors.white); c.setFont("Helvetica-Bold",11) topics = [ "Normal Anatomy & Systematic Approach", "Pneumonia (Lobar, Broncho, Atypical, Aspiration, COVID-19)", "Tuberculosis (Primary, Cavitary, Miliary)", "Pleural Effusion • Pneumothorax • Pulmonary Edema / ARDS", "COPD & Emphysema • Lung Cancer & Masses", "Interstitial Lung Disease • Sarcoidosis • Pulmonary Embolism", "Thoracic Trauma: Rib Fractures, Haemothorax, Aortic Injury", "Quick Reference: Signs, Mnemonics & Differentials", ] y = H*0.41 for t in topics: c.setFillColor(colors.HexColor("#AACCEE")) c.drawCentredString(W/2, y, t); y -= 0.043*H c.setFillColor(GOLD); c.setFont("Helvetica-Bold",9) c.drawCentredString(W/2,2.5*cm,"Sources: Murray & Nadel's Respiratory Medicine • Grainger & Allison's Diagnostic Radiology") c.drawCentredString(W/2,1.8*cm,"Goldman-Cecil Medicine • Rosen's Emergency Medicine • Harrison's 22nd Ed.") c.setFillColor(colors.white); c.setFont("Helvetica",8) c.drawCentredString(W/2,1.1*cm,"Compiled June 2026 | Educational Reference Only") c.restoreState() def later_page(c, doc): c.saveState() c.setFillColor(DARK_BLUE); c.rect(0,0,W,1.1*cm,fill=1,stroke=0) c.setFillColor(colors.white); c.setFont("Helvetica",8) c.drawCentredString(W/2, 0.38*cm, f"CXR Clinical Handbook | Pulmonary Department | Page {doc.page}") c.setFillColor(GOLD); c.setFont("Helvetica-Bold",8) c.drawString(1.5*cm,0.38*cm,"INTERN EDITION 2026") c.restoreState() # ── STORY ───────────────────────────────────────────────────────────────────── story = [Spacer(1, H)] # cover placeholder # ── TOC ─────────────────────────────────────────────────────────────────────── story.append(PageBreak()) story.append(hdr("TABLE OF CONTENTS", "", DARK_BLUE)) story.append(SP(0.5)) toc = [ ("1", MED_BLUE, "How to Read a CXR", "ABCDE approach, normal anatomy, technical adequacy"), ("2", RED, "Pneumonia", "Lobar, bronchopneumonia, atypical, aspiration, COVID-19"), ("3", TEAL, "Tuberculosis", "Primary, reactivation, miliary, HIV-associated"), ("4", PURPLE, "Pleural Effusion", "Transudate, exudate, massive, loculated"), ("5", ORANGE, "Pneumothorax", "Spontaneous, tension, traumatic, iatrogenic"), ("6", GREEN, "Pulmonary Edema & ARDS", "Cardiogenic vs non-cardiogenic, Kerley lines, Berlin criteria"), ("7", GOLD, "COPD & Emphysema", "Hyperinflation, bullae, barrel chest, cor pulmonale"), ("8", PURPLE, "Lung Cancer & Masses", "SPN, Fleischner, Pancoast, cannon-ball mets"), ("9", MED_BLUE, "Interstitial Lung Disease", "UIP, NSIP, sarcoidosis, asbestosis, silicosis"), ("10",RED, "Pulmonary Embolism", "Westermark, Hampton's Hump, Palla's sign"), ("11",colors.HexColor("#922B21"), "Thoracic Trauma", "Rib fractures, haemothorax, contusion, aortic injury"), ("12",DARK_BLUE, "Other Conditions", "Atelectasis, bronchiectasis, mediastinal masses, abscess"), ("13",DARK_BLUE, "Quick Reference", "Classic signs, mnemonics, white-hemithorax differential"), ] for num, col, title, sub in toc: row = [[p(f"<b><font color='white'>{num}</font></b>",sb), p(f"<b>{title}</b><br/><font size='8' color='grey'>{sub}</font>",sb)]] t = Table(row, colWidths=[1.2*cm, W-5.2*cm]) t.setStyle(TableStyle([ ('BACKGROUND',(0,0),(0,0),col), ('BACKGROUND',(1,0),(1,0),LIGHT_GREY), ('TOPPADDING',(0,0),(-1,-1),5),('BOTTOMPADDING',(0,0),(-1,-1),5), ('LEFTPADDING',(0,0),(-1,-1),8),('RIGHTPADDING',(0,0),(-1,-1),8), ('BOX',(0,0),(-1,-1),0.5,col),('VALIGN',(0,0),(-1,-1),'MIDDLE'), ])) story += [t, Spacer(1,2)] # ════════════════════ SECTION 1: HOW TO READ ══════════════════════════════════ story.append(PageBreak()) story.append(hdr("SECTION 1","How to Read a Chest X-Ray — Systematic Approach", MED_BLUE,"1")) story.append(SP()) story.append(p("Every CXR must be assessed systematically. Use the <b>ABCDEF</b> approach every single time — even on a 'normal-looking' film.")) story.append(SP(0.2)) story.append(grid( ["Step","Structure","What to Assess"], [["A","Airway","Trachea midline? Carina angle <70°? ETT/tracheostomy position?"], ["B","Bones & Soft Tissue","Rib fractures, clavicle, vertebrae, subcutaneous emphysema, breast shadows"], ["C","Cardiac","CTR <0.5 on PA? Cardiac borders sharp? Aortic knuckle visible?"], ["D","Diaphragm","Right > left. CP angles sharp? Free air under diaphragm?"], ["E","Everything (lung fields)","Both fields: opacities, hyperinflation, masses, hilar size, vascular markings"], ["F","Foreign bodies / Lines","ET tube, CVC tip, NGT, chest drain, pacemaker — all positions checked?"]], [0.8*cm,3.5*cm,W-8.3*cm], MED_BLUE, LIGHT_BLUE )) story.append(SP(0.5)) story.append(subhdr("Technical Adequacy — Check Before Reading Any CXR", TEAL)) story.append(SP(0.2)) story.append(grid( ["Parameter","Adequate","Pitfall if Inadequate"], [["Rotation","Medial clavicle heads equidistant from spinous process","Mediastinal shift artefact, false cardiomegaly"], ["Inspiration","6 anterior / 10 posterior ribs above diaphragm","Basilar crowding, apparent cardiomegaly, basal opacity"], ["Projection","PA preferred; AP magnifies heart 15-20%","Never calculate CTR on AP film reliably"], ["Exposure","Vertebrae just visible behind cardiac shadow","Over = misses infiltrates; Under = artefactual white-out"]], [2.5*cm,5*cm,6.5*cm], TEAL, LIGHT_TEAL )) story.append(SP(0.5)) story.append(subhdr("Normal Landmarks", MED_BLUE)) story.append(SP(0.2)) for b in [ "<b>Trachea:</b> Midline, slight right deviation at aortic arch is normal", "<b>Carina:</b> Angle normally <70°. Wide carina = left atrial enlargement", "<b>Right heart border:</b> SVC (upper) + right atrium (lower). Obliteration = RML consolidation (silhouette sign)", "<b>Left heart border:</b> Aortic knuckle + main pulmonary artery + left ventricle. Obliteration = lingular/LUL pathology", "<b>Hila:</b> Left normally higher than right. Hila = pulmonary arteries + veins + bronchi + lymph nodes", "<b>Cardiothoracic ratio (CTR):</b> <0.5 on erect PA. Heart width / widest chest width", "<b>Costophrenic angles:</b> Normally sharp. Blunting = ≥200 mL pleural fluid", "<b>Lung markings:</b> Should reach the periphery. Absent = PTX or severe emphysema/bullae", ]: story.append(bu(b)) # ════════════════════ SECTION 2: PNEUMONIA ════════════════════════════════════ story.append(PageBreak()) story.append(hdr("SECTION 2","Pneumonia — All Patterns","#C0392B","2")) story.append(SP()) story.append(p("Pneumonia is the most common infectious cause of death globally. The CXR pattern guides likely aetiology and antibiotic choice.")) story.append(SP()) story.append(subhdr("2.1 Lobar Pneumonia — Streptococcus pneumoniae / Klebsiella", RED)) story.append(SP(0.2)) for x in img_cap("lobar_pna", "Lobar Pneumonia (Pneumococcal) — Homogeneous right upper lobe consolidation conforming to lobar shape, with air bronchograms (arrows). This pattern is classic for Streptococcus pneumoniae or Klebsiella pneumoniae. (Murray & Nadel's Respiratory Medicine)", 11*cm, 8*cm): story.append(x) story.append(SP(0.2)) story.append(two_col( ["Homogeneous lobar/segmental opacity", "Air bronchograms (patent airways in consolidated lung)", "Sharp margin at fissure boundary", "Bulging fissure sign (Klebsiella)", "Silhouette sign — loss of adjacent border", "Volume preserved or slightly increased", "Pleural effusion in 25-50%"], ["Strep pneumoniae: most common CAP organism", "Klebsiella: upper lobe, cavitation, bulging fissure, alcoholics", "Air bronchogram = consolidation, not collapse", "Cavitation: thick-walled = Klebsiella, Staph, anaerobes", "Treat: beta-lactam ± macrolide (CAP)", "Legionella: check urinary antigen, rapid progression", "No improvement at 48-72h = consider resistant organism or empyema"], "CXR Signs","Clinical Significance",RED )) story.append(SP()) story.append(subhdr("2.2 Bronchopneumonia — Staphylococcus, Klebsiella, H. influenzae", RED)) story.append(SP(0.2)) for x in img_cap("klebsiella", "Bronchopneumonia (Klebsiella) — Patchy right lower lobe nodular opacities (arrows). CT (B-E) confirms centrilobular nodules following bronchial tree distribution — hallmark of bronchopneumonia. (Murray & Nadel's Respiratory Medicine)", 13*cm, 8*cm): story.append(x) story.append(SP(0.2)) for b in [ "<b>Pattern:</b> Patchy, multifocal, bilateral, peribronchial distribution — follows bronchial tree", "<b>Bilateral lower lobe predominance</b> — inhomogeneous, non-lobar", "Air bronchograms usually absent; opacity less dense than lobar pneumonia", "<b>Cavitation:</b> 10-30% with Klebsiella, Staph aureus → lung abscess with air-fluid level", "Pleural empyema: think Staph aureus, anaerobes — needs drainage + prolonged antibiotics", ]: story.append(bu(b)) story.append(SP()) story.append(subhdr("2.3 Lung Abscess", RED)) story.append(SP(0.2)) for x in img_cap("abscess","Lung Abscess — Subpleural right apical thick-walled cavity with air-fluid level (arrows). Caused by Klebsiella, Staph, anaerobes or aspiration. CT (B,C) confirms cavity. (Murray & Nadel's Respiratory Medicine)", 12*cm, 7*cm): story.append(x) story.append(SP()) story.append(subhdr("2.4 Atypical Pneumonia — Mycoplasma, Legionella, Viruses", RED)) story.append(SP(0.2)) for x in img_cap("atypical_pna","Atypical Pneumonia (Adenovirus) — Bilateral perihilar peribronchial thickening. Interstitial infiltrates typical of viral/atypical infection. CXR looks worse than patient clinically — classic 'walking pneumonia'. (Murray & Nadel's Respiratory Medicine)", 10*cm, 7*cm): story.append(x) story.append(SP(0.2)) for b in [ "<b>Interstitial pattern:</b> Reticular/reticulonodular infiltrates, peribronchial thickening", "<b>Bilateral perihilar distribution</b> — 'bat-wing' in severe cases", "<b>Mycoplasma:</b> Lower lobe, can look lobar. Cold agglutinins positive. Treat: macrolide/doxycycline", "<b>Legionella:</b> Rapidly progressive consolidation, pleural effusion 50%, hyponatraemia, diarrhoea", "<b>PCP (Pneumocystis):</b> Bilateral perihilar GGO in immunocompromised — CD4 <200", ]: story.append(bu(b)) story.append(SP()) story.append(subhdr("2.5 Aspiration Pneumonia", RED)) story.append(SP(0.2)) for x in img_cap("aspiration","Aspiration Pneumonia — RLL consolidation with volume loss + tracheal deviation (right). CT (B-E): dependent consolidation with centrilobular nodules. Dependent segment location is the key diagnostic clue. (Murray & Nadel's Respiratory Medicine)", 13*cm, 8*cm): story.append(x) story.append(SP(0.2)) for b in [ "<b>Location:</b> Dependent segments — RLL when upright; posterior segments when supine", "Right > left because right main bronchus is more vertical", "Complications: lung abscess, empyema, ARDS", "Risk factors: stroke, dysphagia, altered consciousness, alcoholism, NG tube", ]: story.append(bu(b)) story.append(SP()) story.append(subhdr("2.6 COVID-19 Pneumonia", RED)) story.append(SP(0.2)) for x in img_cap("covid_early","COVID-19 (Early — Exudative Phase): Bilateral peripheral upper lobe rounded ground-glass opacities. Classic early appearance, days 1-7. (Goldman-Cecil Medicine)", 10*cm, 7*cm): story.append(x) story.append(SP(0.2)) for x in img_cap("covid_late","COVID-19 (Organising Phase): Multifocal confluent consolidation + GGO bilaterally. Small left pleural effusion. Days 7-14 of illness. (Goldman-Cecil Medicine)", 10*cm, 7*cm): story.append(x) story.append(SP(0.2)) story.append(grid(["Phase","Timing","CXR Features"], [["Early (exudative)","Days 1-7","Bilateral peripheral GGO, often upper lobe, rounded"], ["Organising","Days 7-14","Consolidation + GGO, multifocal, bilateral, peripheral > central"], ["Advanced / ARDS","Day 14+","Diffuse bilateral consolidation, white-out, pleural effusion"], ["CT only","Any stage","Crazy paving, reversed halo, traction bronchiectasis, fibrosis"]], [3*cm, 2*cm, 9*cm], RED, LIGHT_RED)) # ════════════════════ SECTION 3: TB ═══════════════════════════════════════════ story.append(PageBreak()) story.append(hdr("SECTION 3","Tuberculosis — Primary, Reactivation, Miliary", "#1B7A6E","3")) story.append(SP()) story.append(p("TB is the world's leading infectious disease killer. CXR pattern varies critically with the patient's immune status and phase of infection.")) story.append(SP()) for x in img_cap("tb_primary","Primary TB — RLL consolidation (Ghon focus) + right hilar lymphadenopathy (arrows) = Ghon Complex. This is the classic primary TB appearance in children and young adults. (Murray & Nadel's Respiratory Medicine)", 10*cm, 7*cm): story.append(x) story.append(SP(0.2)) story.append(grid(["Type","Distribution","CXR Features","Cavitation"], [["Primary","Middle/lower zone, any","Consolidation + hilar LN (Ghon complex), pleural effusion","Rare"], ["Reactivation","Apical + posterior upper lobes","Infiltrate, fibrosis, calcification, cavitation, bronchogenic spread","Common ~50%"], ["Miliary","Diffuse bilateral","1-3mm uniform nodules ('millet seeds') everywhere","No"], ["HIV-associated","Lower zones, diffuse","Atypical — lower lobe, LN, effusions, no cavitation","Uncommon"], ["EPTB","No lung finding","Normal CXR possible — sputum AFB negative; IGRA/biopsy needed","N/A"]], [2.5*cm,3.5*cm,6*cm,2*cm], TEAL, LIGHT_TEAL)) story.append(SP()) for x in img_cap("tb_cavitary","Cavitary (Reactivation) TB — Extensive right upper lobe thick-walled cavitation. Most common in apical/posterior RUL and apical-posterior LUL. Send sputum AFB ×3 + GeneXpert immediately. (Murray & Nadel's Respiratory Medicine)", 10*cm, 7*cm): story.append(x) story.append(SP()) for x in img_cap("tb_miliary","Miliary TB — Innumerable tiny 1-3mm randomly distributed nodules both lungs. Results from haematogenous dissemination. Screen for CNS TB (LP). Mortality high if untreated. (Murray & Nadel's Respiratory Medicine)", 10*cm, 7*cm): story.append(x) story.append(SP(0.3)) story.append(box([ p("<b>CXR Red Flags for TB — Act Immediately:</b>",sb), bu("Upper lobe opacity or cavity in any adult, especially apical/posterior segment"), bu("Bilateral upper lobe infiltrates with or without cavitation"), bu("Miliary pattern — screen for meningitis urgently; start empirical treatment"), bu("Persistent opacity >2-3 weeks not responding to standard antibiotics"), bu("Pleural effusion in young person from TB-endemic region"), bu("Lower lobe disease + lymphadenopathy in HIV patient = atypical TB"), ], LIGHT_TEAL, TEAL)) # ════════════════════ SECTION 4: PLEURAL EFFUSION ═════════════════════════════ story.append(PageBreak()) story.append(hdr("SECTION 4","Pleural Effusion", "#6C3483","4")) story.append(SP()) for x in img_cap("effusion","Pleural Effusion — CXR shows meniscus sign (right): homogeneous basal opacity, concave upper border, higher laterally. CT (left) shows full extent of fluid and distinguishes free from loculated. (Grainger & Allison's Diagnostic Radiology)", 12*cm, 8*cm): story.append(x) story.append(SP(0.2)) story.append(grid(["Volume","CXR Finding","Significance"], [["50-75 mL","Posterior CP blunting on lateral only","Visible only on lateral decubitus or US"], ["200-500 mL","Lateral CP angle blunted on PA","Earliest sign on standard PA CXR"], ["500-1000 mL","Meniscus sign — concave upper border, higher laterally","Classical moderate effusion"], [">1000 mL","Opacification >4th anterior rib; mediastinal shift away","Large effusion"], [">2000 mL","Complete hemithoracic opacification; contralateral shift","Massive effusion"]], [2*cm,7*cm,5*cm], PURPLE, LIGHT_PURPLE)) story.append(SP()) story.append(two_col( ["Meniscus sign — free fluid", "No shift with large effusion = mesothelioma or collapse", "Subpulmonic effusion: high hemidiaphragm, peak lateral", "Loculated: fixed, D-shaped, doesn't move with position", "Bilateral small effusions: cardiac failure (commonest)"], ["Transudates: CHF, cirrhosis, nephrotic syndrome, PE", "Exudates (Light's criteria): LDH ratio >0.6, protein ratio >0.5", "Causes of exudate: pneumonia/parapneumonic, TB, malignancy, PE", "Haemothorax: blood — trauma, aortic rupture, malignancy", "Chylothorax: milky fluid, triglycerides >1.24 mmol/L, trauma/lymphoma"], "CXR Signs","Aetiology", PURPLE)) # ════════════════════ SECTION 5: PNEUMOTHORAX ═════════════════════════════════ story.append(PageBreak()) story.append(hdr("SECTION 5","Pneumothorax — Spontaneous, Tension, Traumatic","#E67E22","5")) story.append(SP()) story.append(p("<b>TENSION PNEUMOTHORAX IS A CLINICAL DIAGNOSIS — DO NOT WAIT FOR CXR. TREAT IMMEDIATELY.</b>")) story.append(SP()) p1 = fetch(IMGS["ptx_insp"], 7*cm, 7*cm) p2 = fetch(IMGS["ptx_exp"], 7*cm, 7*cm) if p1 and p2: t_ptx = Table([[p1, p2]], colWidths=[7.5*cm,7.5*cm]) t_ptx.setStyle(TableStyle([('ALIGN',(0,0),(-1,-1),'CENTER'),('VALIGN',(0,0),(-1,-1),'TOP')])) story.append(t_ptx) story.append(p("Left Spontaneous Pneumothorax: (A) Inspiration — visceral pleural line, absent peripheral lung markings. (B) Expiration — PTX becomes MORE apparent. Use expiratory film to confirm small PTX. (Grainger & Allison)", sc)) story.append(SP(0.2)) story.append(box([ p("<b>TENSION PNEUMOTHORAX — Emergency Action:</b>",sb), bu("Trachea deviated AWAY from the affected side"), bu("Complete ipsilateral lung collapse + contralateral mediastinal shift"), bu("Depressed ipsilateral hemidiaphragm"), bu("Absent breath sounds + hypotension + tachycardia + hypoxia = NEEDLE DECOMPRESSION NOW"), bu("Technique: 2nd ICS, mid-clavicular line, 14G cannula — immediate gush of air confirms"), bu("Follow with formal chest drain (5th ICS, anterior axillary line)"), ], LIGHT_RED, RED)) story.append(SP()) story.append(grid(["Type","CXR Features","Management"], [["Primary spontaneous","Visceral pleural line, avascular zone peripherally, apex collapse","Small (<2cm): O2 + observe. Large: aspiration or drain"], ["Secondary spontaneous","As above on background COPD/asthma/fibrosis/bullae","All require drainage — poorly tolerated"], ["Tension","Complete collapse, trachea deviates away, shift, depressed diaphragm","IMMEDIATE needle decompression + drain"], ["Traumatic","Rib fractures ± haemothorax, subcutaneous emphysema","Drain + treat underlying injuries"], ["Iatrogenic","Post-CVC, post-biopsy, post-intubation. Always check post-procedure CXR","Aspiration usually sufficient for small"], ["Occult","Absent on CXR, present on CT — 'deep sulcus sign' on supine AP","Observe or drain depending on size/ventilation status"]], [3*cm,6*cm,5*cm], ORANGE, LIGHT_ORANGE)) story.append(SP()) for x in img_cap("ptx_skinfold","PITFALL — Skin Fold Mimicking Right Pneumothorax (arrows). Features distinguishing skin fold: line extends BEYOND chest wall, blood vessels visible lateral to line, line orientation inconsistent with collapsed lung. Always look for these before labelling a PTX. (Grainger & Allison)", 10*cm, 7*cm): story.append(x) # ════════════════════ SECTION 6: PULMONARY EDEMA ══════════════════════════════ story.append(PageBreak()) story.append(hdr("SECTION 6","Pulmonary Edema & ARDS","#1E8449","6")) story.append(SP()) story.append(p("Distinguishing cardiogenic from non-cardiogenic pulmonary oedema on CXR is one of the most clinically important interpretations you will make.")) story.append(SP()) for x in img_cap("edema","Pulmonary Oedema (HAPE) — (A) Bilateral patchy alveolar infiltrates. (B) Near-complete resolution after oxygen + nifedipine. Note: absence of cardiomegaly = non-cardiogenic. (Rosen's Emergency Medicine)", 12*cm, 8*cm): story.append(x) story.append(SP(0.2)) story.append(grid(["CXR Feature","Cardiogenic Oedema","Non-cardiogenic (ARDS)"], [["Cardiac size","Cardiomegaly (CTR >0.5)","Normal heart size"], ["Distribution","Perihilar 'bat-wing', central","Peripheral/diffuse, patchy"], ["Kerley B-lines","Present — short horizontal lines at bases","Absent"], ["Pleural effusions","Bilateral, common","Rare or small"], ["Vascular pattern","Upper lobe diversion","Normal vascularity"], ["Onset","Hours (flash)","12-24h after trigger"], ["Berlin ARDS","Not ARDS by definition","P/F <300, bilateral, no CHF"]], [4*cm,5*cm,5*cm], GREEN, LIGHT_GREEN)) story.append(SP()) story.append(box([ p("<b>Interstitial Oedema — Classic CXR Signs:</b>",sb), bu("<b>Kerley A-lines:</b> Long diagonal lines from hilum to periphery — dilated lymphatics"), bu("<b>Kerley B-lines:</b> Short (1-3cm) horizontal lines at lung bases, perpendicular to pleura — thickened interlobular septa"), bu("<b>Peribronchial cuffing:</b> Thickened bronchial walls seen end-on as 'ring shadows' >1cm from hilum"), bu("<b>Perihilar haziness:</b> Blurring of hilar vascular markings"), bu("<b>Upper lobe diversion:</b> Upper lobe veins larger than lower lobe — left atrial pressure >12 mmHg"), bu("<b>Alveolar oedema:</b> Bat-wing perihilar consolidation, air bronchograms — LVEDP >25 mmHg"), ], LIGHT_GREEN, GREEN)) # ════════════════════ SECTION 7: COPD ═════════════════════════════════════════ story.append(PageBreak()) story.append(hdr("SECTION 7","COPD & Emphysema","#7D6608","7")) story.append(SP()) story.append(p("COPD is the 3rd leading cause of death globally. CXR has low sensitivity for early disease but classic features appear in moderate-severe COPD.")) story.append(SP()) story.append(grid(["CXR Sign","Description","What It Means"], [["Hyperinflation",">6 anterior / >10 posterior ribs above diaphragm","Air trapping — obstructive physiology"], ["Flattened diaphragms","Loss of diaphragmatic dome; below anterior 7th rib","Severe air trapping"], ["Increased AP diameter","'Barrel chest' on lateral — AP approaches transverse diameter","Chronic overinflation"], ["Increased retrosternal space",">2.5cm between sternum and aorta on lateral","Emphysema"], ["Bullae","Thin-walled air spaces >1cm, avascular, upper lobes","Severe emphysema — risk of PTX"], ["Attenuated vasculature","Reduced peripheral markings — 'pruned tree'","Vascular destruction"], ["Narrow/vertical heart","'Hanging heart' — elongated cardiac silhouette","Low flat diaphragm pushes heart down"], ["Enlarged hila","Prominent central pulmonary arteries","Pulmonary hypertension (cor pulmonale)"], ["Cor pulmonale","Right heart enlargement, elevated right diaphragm","End-stage COPD complication"]], [4*cm,6*cm,4*cm], colors.HexColor("#7D6608"), LIGHT_GOLD)) # ════════════════════ SECTION 8: LUNG CANCER ══════════════════════════════════ story.append(PageBreak()) story.append(hdr("SECTION 8","Lung Cancer & Pulmonary Masses","#6C3483","8")) story.append(SP()) story.append(p("Lung cancer is the leading cause of cancer death worldwide. Any new or enlarging opacity in an adult smoker must be investigated as malignancy until proven otherwise.")) story.append(SP()) story.append(grid(["Type","Location","CXR Appearance","Key Feature"], [["Squamous cell","Central, perihilar","Hilar/perihilar mass, obstructive collapse, cavitation","Cavitation 20%"], ["Adenocarcinoma","Peripheral","Spiculated peripheral mass/nodule, pleural tag, ground-glass halo","Most common overall"], ["Small cell (SCLC)","Central","Bulky mediastinal/hilar LN, 'potato' hilum, rapid growth","Paraneoplastic syndromes"], ["Large cell","Peripheral","Large (>4cm) irregular mass, necrosis possible","Diagnosis of exclusion"], ["Pancoast (superior sulcus)","Apex","Apical mass ± rib 1-3 destruction, Horner syndrome","T4 — requires MRI"], ["Carcinoid","Central/peripheral","Well-defined nodule, may calcify; obstructive features if central","Low-grade malignancy"], ["Metastases","Bilateral peripheral","Multiple 'cannon-ball' rounded nodules — variable sizes","Breast, colon, renal, melanoma"]], [3*cm,2.5*cm,5.5*cm,3*cm], PURPLE, LIGHT_PURPLE)) story.append(SP()) story.append(box([ p("<b>Solitary Pulmonary Nodule (SPN) — Fleischner Society Approach:</b>",sb), bu("<b>Benign features:</b> Smooth margins, <6mm, central/laminar/popcorn calcification, stable ≥2 years"), bu("<b>Malignant features:</b> Spiculated/lobulated, >8mm, growing, part-solid, upper lobe, smoker >35 pack-years"), bu("<b><6mm, low risk:</b> No CT follow-up routinely"), bu("<b>6-8mm, low risk:</b> CT at 6-12 months; then at 18-24 months"), bu("<b>>8mm:</b> CT at 3 months, PET-CT, consider biopsy"), bu("<b>Any growth:</b> Urgently refer regardless of size"), ], LIGHT_PURPLE, PURPLE)) story.append(SP()) story.append(box([ p("<b>Signs of Malignancy on CXR — Do Not Miss:</b>",sb), bu("Hilar enlargement — bilateral = sarcoid/lymphoma; unilateral = bronchogenic carcinoma/mets"), bu("Raised hemidiaphragm — phrenic nerve palsy from apical/mediastinal malignancy"), bu("Rib destruction + apical mass = Pancoast tumour (MRI spine urgently)"), bu("Obstructive collapse — right upper lobe collapse raises right hilum (reverse S of Golden)"), bu("Pleural effusion + lung mass = likely malignant effusion, refer respiratory oncology"), ], LIGHT_PURPLE, PURPLE)) # ════════════════════ SECTION 9: ILD ══════════════════════════════════════════ story.append(PageBreak()) story.append(hdr("SECTION 9","Interstitial Lung Disease & Sarcoidosis","#1A5276","9")) story.append(SP()) story.append(grid(["Pattern","CXR Appearance","Common Causes"], [["UIP/IPF","Bilateral basal subpleural reticular pattern, honeycombing, traction bronchiectasis","Idiopathic pulmonary fibrosis (irreversible)"], ["NSIP","Bilateral GGO, lower lobe, peribronchovascular. Less honeycombing","CTD-ILD, drugs, hypersensitivity"], ["COP","Bilateral peripheral consolidation, lower lobe, reversed halo sign","Cryptogenic organising pneumonia"], ["Sarcoidosis","Bilateral hilar LN ± upper lobe reticulonodular opacities","Young adults, ACE elevated"], ["HP","Bilateral upper/mid nodules, GGO, may spare bases","Farmer's lung, bird-fancier's"], ["Asbestosis","Bilateral basal fibrosis + pleural plaques/calcification","Occupational asbestos exposure"], ["Silicosis","Upper lobe nodules, eggshell hilar node calcification, PMF","Miners, sandblasters"], ["Lymphangitis carcinomatosa","Bilateral Kerley B-lines, nodules, hilar LN — progressive","Breast, stomach, lung mets to lymphatics"]], [3*cm,6.5*cm,4.5*cm], colors.HexColor("#1A5276"), LIGHT_BLUE)) story.append(SP()) story.append(subhdr("Sarcoidosis — Scadding Radiological Staging", MED_BLUE)) story.append(SP(0.2)) story.append(grid(["Stage","CXR Finding","Spontaneous Remission"], [["0","Normal CXR","Possible"], ["I","Bilateral hilar lymphadenopathy (BHL) alone","60-80%"], ["II","BHL + parenchymal infiltrates","40-70%"], ["III","Parenchymal disease only, no BHL","10-20%"], ["IV","Pulmonary fibrosis, honeycombing, traction bronchiectasis","Irreversible"]], [1.5*cm,8*cm,4.5*cm], MED_BLUE, LIGHT_BLUE)) # ════════════════════ SECTION 10: PE ══════════════════════════════════════════ story.append(PageBreak()) story.append(hdr("SECTION 10","Pulmonary Embolism — CXR Signs","#C0392B","10")) story.append(SP()) story.append(p("<b>Key point: CXR is normal in up to 40% of PE cases.</b> Its main role is to exclude other causes of breathlessness (PTX, pneumonia, pulmonary oedema). A <b>normal CXR in a hypoxic patient</b> should immediately raise suspicion for PE. CTPA is diagnostic standard.")) story.append(SP()) story.append(grid(["CXR Sign","Appearance","Sensitivity"], [["Normal CXR","Most common — present in ~40% of PE","N/A — unhelpful diagnostically"], ["Westermark sign","Focal oligaemia (lucency) distal to clot — 'pruned' vessels","10-15% (low)"], ["Hampton's Hump","Peripheral wedge-shaped pleural-based opacity (infarction), flat base at pleura","10-15% (low)"], ["Palla's sign","Enlarged right descending pulmonary artery","Low"], ["Fleischner sign","Prominent central pulmonary artery — clot distension","Low"], ["Elevated hemidiaphragm","Ipsilateral diaphragm elevation from reduced lung volume","Moderate"], ["Platelike atelectasis","Horizontal linear atelectasis at lung bases — very common","Common but non-specific"], ["Small pleural effusion","Ipsilateral haemorrhagic effusion with infarction","Moderate"]], [3.5*cm,6.5*cm,4*cm], RED, LIGHT_RED)) story.append(SP()) story.append(box([ p("<b>Wells Score — Pre-test Probability for PE:</b>",sb), bu("Clinical signs/symptoms of DVT: +3"), bu("PE more likely than alternative diagnosis: +3"), bu("Heart rate >100: +1.5"), bu("Immobilisation/surgery in past 4 weeks: +1.5"), bu("Previous PE or DVT: +1.5"), bu("Haemoptysis: +1"), bu("Malignancy (treatment within 6 months): +1"), bu("<b>Score <2 = low (PE unlikely); 2-6 = moderate; >6 = high (PE likely)</b>"), bu("Low probability + negative D-dimer = PE excluded. High probability = CTPA urgently"), ], LIGHT_RED, RED)) # ════════════════════ SECTION 11: TRAUMA ══════════════════════════════════════ story.append(PageBreak()) story.append(hdr("SECTION 11","Thoracic Trauma — Emergency CXR Interpretation","#922B21","11")) story.append(SP()) story.append(p("Thoracic trauma causes 25% of trauma deaths. Portable AP CXR is the first-line image in the trauma bay. These findings are time-critical — recognise and act immediately.")) story.append(SP()) story.append(grid(["Injury","CXR Findings","Urgency & Action"], [["Rib Fractures","Cortical break/step. Flail chest: ≥2 fractures ≥3 consecutive ribs. Check for underlying PTX/haemothorax","Analgesia, physio. Flail → ICU, possible intubation"], ["Haemothorax","Homogeneous opacity (blunted CP angle). Massive: complete hemithoracic opacification","Chest drain (large bore 32-36Fr). Massive (>1.5L): thoracotomy"], ["Pulmonary Contusion","Patchy/confluent consolidation, no lobar boundaries, appears within hours, worsens 24-48h","Supportive — avoid fluid overload; risk of progression to ARDS"], ["Traumatic Pneumothorax","Visceral pleural line; 'Deep sulcus sign' on supine AP (deep lucent CP angle)","Chest drain. Tension: needle decompression immediately"], ["Aortic Injury","Widened mediastinum >8cm, loss of aortic knuckle, left apical cap, trachea right, NG tube right","URGENT CT aorta — Vascular surgery NOW"], ["Diaphragm Rupture","Elevated hemidiaphragm, bowel loops in chest, NG tube in chest","Surgical repair; left-sided more common (70%)"], ["Tracheobronchial Injury","Pneumomediastinum, subcutaneous emphysema, persistent PTX despite drain","Bronchoscopy + surgical repair — high mortality"], ["Cardiac Tamponade","Globular 'water-bottle' heart, clear lungs, no pulmonary oedema","Emergency pericardiocentesis/pericardiotomy"], ["Oesophageal Rupture","Pneumomediastinum + left effusion + SC emphysema (Boerhaave)","Emergency surgery — extreme mortality if delayed"]], [3.5*cm,6*cm,4.5*cm], colors.HexColor("#922B21"), LIGHT_RED)) story.append(SP()) story.append(box([ p("<b>Widened Mediastinum — Checklist for Aortic Injury (Any 2 = CT Aorta Urgently):</b>",sb), bu("Mediastinum >8 cm at aortic knuckle on AP CXR"), bu("Loss of normal aortic contour — indistinct aortic knuckle"), bu("Left apical pleural cap (haematoma tracking into apex)"), bu("Trachea deviated to the RIGHT"), bu("Nasogastric tube deviated to the RIGHT"), bu("Depression of left main bronchus >40 degrees below horizontal"), bu("First/second rib fractures — high-energy trauma marker"), ], LIGHT_RED, colors.HexColor("#922B21"))) # ════════════════════ SECTION 12: OTHER CONDITIONS ════════════════════════════ story.append(PageBreak()) story.append(hdr("SECTION 12","Other Pulmonary Conditions",DARK_BLUE,"12")) story.append(SP()) story.append(grid(["Condition","CXR Appearance","Key Notes"], [["Atelectasis — Lobar","Volume loss, shifted fissures/mediastinum, increased density, elevated diaphragm","RML: loss of right heart border. LLL: loss of left hemidiaphragm"], ["Bronchiectasis","Tram-track lines, ring shadows, cystic spaces, mucus plugging ('gloved finger')","CF, PCD, immunodeficiency, post-infection"], ["Lung Abscess","Thick-walled cavity with air-fluid level — dependent segments","Anaerobes, Klebsiella, Staph, aspiration; prolonged antibiotics"], ["Cardiomegaly","CTR >0.5 PA. Globular = pericardial effusion. Boot-shaped = ToF","Dilated, hypertrophic cardiomyopathy, pericardial disease"], ["Mediastinal Masses","Ant: 4T's (Thymoma/Teratoma/Thyroid/T-lymphoma). Mid: Lymphoma. Post: Neurogenic","CT mandatory for characterisation; position on lateral CXR key"], ["Pleural Plaques","Bilateral pleural thickening/calcification (diaphragm, pericardium)","Pathognomonic of asbestos — marker for mesothelioma risk"], ["Pulmonary Hypertension","Enlarged main PA (>3.2cm), pruned tree periphery, right heart enlargement","COPD, ILD, chronic PE, CTD, idiopathic"], ["Cystic Fibrosis","Upper lobe bronchiectasis, peribronchial thickening, mucus plugging, hyperinflation","Progressive upper lobe fibrosis; Pseudomonas colonisation"], ["Pulmonary AVM","Serpiginous/rounded opacity with visible feeding and draining vessels","HHT; paradoxical embolism risk; CT angiography to confirm"]], [3.5*cm,6*cm,4.5*cm], DARK_BLUE, LIGHT_BLUE)) # ════════════════════ SECTION 13: QUICK REFERENCE ═════════════════════════════ story.append(PageBreak()) story.append(hdr("SECTION 13","Quick Reference — Signs, Mnemonics & Differentials",DARK_BLUE,"13")) story.append(SP()) story.append(subhdr("Classic CXR Signs — Memorise These", DARK_BLUE)) story.append(SP(0.2)) story.append(grid(["Sign","Appearance","Diagnosis"], [["Air bronchogram","Air-filled bronchi visible within opacified lung","Consolidation (pneumonia, ARDS, oedema)"], ["Silhouette sign","Loss of normal border between adjacent structures","Localises pathology: RML = right heart border; LLL = left diaphragm"], ["Meniscus sign","Concave upper opacity border, higher laterally","Free pleural effusion"], ["Hampton's Hump","Peripheral pleural-based wedge opacity","Pulmonary infarction (PE)"], ["Westermark sign","Focal oligaemia / avascular zone","Pulmonary embolism"], ["Bulging fissure","Fissure bowed away from consolidated lobe","Klebsiella, severe lobar pneumonia"], ["Deep sulcus sign","Deep lucent CP angle on supine AP film","Pneumothorax (supine patient)"], ["Bat-wing pattern","Bilateral perihilar alveolar opacification","Cardiogenic pulmonary oedema"], ["Eggshell calcification","Ring calcification of hilar lymph nodes","Silicosis (pathognomonic), sarcoidosis"], ["Water-bottle heart","Globular enlarged cardiac silhouette","Pericardial effusion / tamponade"], ["Sail sign","Triangular thymic opacity in child","Normal thymus — not consolidation"], ["Continuous diaphragm","Air visible beneath both hemidiaphragms","Pneumomediastinum"], ["Kerley B-lines","Short horizontal peripheral basal lines","Pulmonary oedema, lymphangitis carcinomatosa"], ["Golden S sign","S-shaped curve of right minor fissure","RUL collapse + central hilar mass (lung cancer)"], ["Crazy paving (CT)","GGO + superimposed interlobular thickening","COVID-19, PAP, ARDS, PCP, lipoid pneumonia"]], [4*cm,5*cm,5*cm], DARK_BLUE, LIGHT_BLUE)) story.append(SP()) story.append(box([ p("<b>Essential Mnemonics for the Intern:</b>",sb), bu("<b>ABCDEF</b> — Airway, Bones, Cardiac, Diaphragm, Everything (lung fields), Foreign bodies/Lines"), bu("<b>4T's (anterior mediastinum)</b> — Thymoma, Teratoma/germ cell, Thyroid, Terrible lymphoma"), bu("<b>CAVITY (causes of cavitation)</b> — Cancer, Abscess, Vasculitis (GPA), Infarct, TB/fungus, Young (congenital)"), bu("<b>BATS (bilateral hilar enlargement)</b> — Berylliosis, Amyloid, TB/lymphoma, Sarcoidosis"), bu("<b>CHEST (effusion causes)</b> — Cardiac, Hepatic, Empyema, Subphrenic, TB/Tumour"), bu("<b>SPLAT (hilar enlargement)</b> — Sarcoid, Primary TB, Lymphoma, Artery (PAH), Tumour (bronchogenic)"), ], LIGHT_GOLD, GOLD)) story.append(SP()) story.append(subhdr("White Hemithorax — Differential Diagnosis", DARK_BLUE)) story.append(SP(0.2)) story.append(grid(["Finding","Mediastinal Shift","Diagnosis"], [["White hemithorax + shift AWAY from opacity","Away","Massive pleural effusion"], ["White hemithorax + shift TOWARD opacity","Toward","Total lung collapse / pneumonectomy"], ["White hemithorax + NO shift","None","Mesothelioma or combined effusion + collapse"], ["Bilateral white-out + no shift","None","Massive pulmonary oedema, bilateral pneumonia, ARDS"]], [5.5*cm,3.5*cm,5*cm], DARK_BLUE, LIGHT_BLUE)) story.append(SP()) story.append(subhdr("Ground-Glass Opacity (GGO) — Differential", MED_BLUE)) story.append(SP(0.2)) story.append(grid(["Pattern","Distribution","Consider"], [["Bilateral, perihilar","Central","Pulmonary oedema (cardiogenic)"], ["Bilateral, peripheral, rounded","Peripheral/upper","COVID-19 (early)"], ["Bilateral, lower lobe","Lower/perihilar","NSIP, drug reaction, hypersensitivity"], ["Unilateral, lobar","Any lobe","Pneumonia (atypical), haemorrhage"], ["Bilateral diffuse","Diffuse","ARDS, PCP (immunocompromised), DAH"], ["With crazy paving","Diffuse","PAP, COVID-19, Pneumocystis, lipoid"]], [4*cm,3.5*cm,6.5*cm], MED_BLUE, LIGHT_BLUE)) # ── REFERENCES ──────────────────────────────────────────────────────────────── story.append(PageBreak()) story.append(hdr("REFERENCES & DISCLAIMER", "", DARK_BLUE)) story.append(SP()) for ref in [ "Murray & Nadel's Textbook of Respiratory Medicine, 7th Ed. — Elsevier 2022 (ISBN 978-0-323-65587-3)", "Grainger & Allison's Diagnostic Radiology, 7th Ed. — Elsevier 2021 (ISBN 978-0-323-76075-1)", "Goldman-Cecil Medicine, 27th Ed. — Elsevier 2024 (ISBN 978-0-323-93034-5)", "Rosen's Emergency Medicine: Concepts & Clinical Practice, 10th Ed.", "Fishman's Pulmonary Diseases and Disorders, 5th Ed.", "Harrison's Principles of Internal Medicine, 22nd Ed. — McGraw Hill 2025", "British Thoracic Society — Pleural Disease, Pneumothorax Guidelines 2023", "Fleischner Society — Pulmonary Nodule Guidelines 2017", "Berlin Definition of ARDS — JAMA 2012;307(23):2526-2533", "WHO Global TB Report 2023", ]: story.append(bu(ref)) story.append(SP(0.5)) story.append(box([ p("<b>Disclaimer:</b> This handbook is an educational quick-reference guide for intern physicians. " "All clinical decisions must be made within the full clinical context and under supervision of " "senior clinicians. CXR findings must be correlated with history, examination, and further imaging. " "Never delay emergency treatment (tension PTX, aortic injury, cardiac tamponade) to obtain CXR.",sb), ], LIGHT_GREY, MID_GREY)) # ── BUILD ───────────────────────────────────────────────────────────────────── doc = SimpleDocTemplate(OUT, pagesize=A4, leftMargin=2*cm, rightMargin=2*cm, topMargin=2*cm, bottomMargin=2*cm, title="Chest X-Ray Clinical Handbook", author="Orris AI Medical Assistant") print("Building PDF...") doc.build(story, onFirstPage=cover_page, onLaterPages=later_page) print(f"SUCCESS: {OUT}") import os; print(f"File size: {os.path.getsize(OUT)/1024:.0f} KB") PYEOF echo "Script written OK"

cd /home/daytona/workspace/cxr-handbook && python3 build.py

cd /home/daytona/workspace/cxr-handbook && sed -i 's/story = \[Spacer(1, H)\]/story = [PageBreak()]/' build.py && python3 build.py

python3 -c " content = open('/home/daytona/workspace/cxr-handbook/build.py').read() content = content.replace('story = [Spacer(1, H)]', 'story = [PageBreak()]') open('/home/daytona/workspace/cxr-handbook/build.py', 'w').write(content) print('Fixed') " cd /home/daytona/workspace/cxr-handbook && python3 build.py

python3 -c " import requests, io from PIL import Image, ImageDraw, ImageFont import os OUT_DIR = '/home/daytona/workspace/cxr-handbook/annotated' os.makedirs(OUT_DIR, exist_ok=True) IMGS = { 'lobar_pneumonia': ('https://cdn.orris.care/cdss_images/4107a235bcb72d12fd7e03be9f1a5dafdd7494a8813a4916ae69a9055048eebd.png', 'Lobar Pneumonia'), 'atypical_pneumonia': ('https://cdn.orris.care/cdss_images/b5c9f16401022bfb2d51e0fdd2637239387d6efed5968b9937a273228a46936e.png', 'Atypical Pneumonia'), 'aspiration': ('https://cdn.orris.care/cdss_images/f97f914e6eb5f10ade11a5d80ffbc4d3088ed5b668ac75125607e5420e132c04.png', 'Aspiration Pneumonia'), 'klebsiella': ('https://cdn.orris.care/cdss_images/3ea63f4fb73b96ca26369c7316999576edae82a8414da54f704a7640fa9a8432.png', 'Bronchopneumonia'), 'covid_early': ('https://cdn.orris.care/cdss_images/9262c9349b671f32e5a28ad645bff170da0f711b5cec9534372b648201998e69.png', 'COVID-19 Early'), 'covid_late': ('https://cdn.orris.care/cdss_images/6825410929b7c2ca0df925a29c2fede605e83e962d14298226fc12afa31c1215.png', 'COVID-19 Late'), 'tb_primary': ('https://cdn.orris.care/cdss_images/a8d4518bc7b3478b3ecea071bce32212f29a89347dd7529573da89d558556a8e.png', 'Primary TB'), 'tb_cavitary': ('https://cdn.orris.care/cdss_images/77ba4b612dfbb87572a123ad3217ba95e9e97f4c32f08a9d5586b4020604a0d3.png', 'Cavitary TB'), 'tb_miliary': ('https://cdn.orris.care/cdss_images/ddac82cec1d3bd545a6e03713c64e9cb5bee36b65f75ee822aa6f0d506c90e97.png', 'Miliary TB'), 'ptx_insp': ('https://cdn.orris.care/cdss_images/3848a612ca4563d4d4b97a4f0ff7d757f1385dcf0fcf6d2d0daf9775e39f3d99.png', 'Pneumothorax'), 'ptx_exp': ('https://cdn.orris.care/cdss_images/85971adc7eb8b4ea0487b6b1b899f501cc0ebc7ef8022e454e23c462b74d33dc.png', 'Pneumothorax Expiration'), 'ptx_skinfold': ('https://cdn.orris.care/cdss_images/e088349db48b2c6f428ba8a29aa3031460df630929623c6504da45a4750c48ac.png', 'Skin Fold vs PTX'), 'effusion': ('https://cdn.orris.care/cdss_images/c67494bd5d0ea8fa634b60d51c5efad65d8de9bdaf7367f31794b662fc55f0ca.png', 'Pleural Effusion'), 'edema': ('https://cdn.orris.care/cdss_images/feeb5dadaec1ba52bb94d97bbc9164a393847675bc322c0e53e5c4adbe4b62a9.png', 'Pulmonary Edema'), 'measles_pna': ('https://cdn.orris.care/cdss_images/8f7e26d9363360744b07202913bff9cc454985502a95b7dfdfc56b53fbd2ac63.png', 'Viral Pneumonia'), 'covid_ct': ('https://cdn.orris.care/cdss_images/275934de74457c5e869a9d557a18aa10e476b1b25c779e34aa542925a000431e.png', 'COVID CT'), } # Annotation definitions: key -> list of (type, coords, color, label) # type: 'arrow' (x1,y1,x2,y2), 'circle' (cx,cy,r), 'rect' (x1,y1,x2,y2) # coords as fractions of image width/height ANNOTS = { 'lobar_pneumonia': [ ('circle', (0.65, 0.35, 0.12), 'red', 'Consolidation'), ('arrow', (0.85, 0.28, 0.72, 0.33), 'red', 'Air bronchogram'), ], 'atypical_pneumonia': [ ('circle', (0.38, 0.40, 0.10), 'yellow', 'Perihilar infiltrate'), ('circle', (0.62, 0.42, 0.10), 'yellow', 'Bilateral infiltrate'), ('arrow', (0.50, 0.20, 0.50, 0.35), 'yellow', 'Peribronchial thickening'), ], 'aspiration': [ ('circle', (0.70, 0.65, 0.12), 'red', 'RLL consolidation'), ('arrow', (0.90, 0.55, 0.76, 0.62), 'red', 'Dependent segment'), ('arrow', (0.45, 0.30, 0.50, 0.40), 'cyan', 'Tracheal deviation'), ], 'klebsiella': [ ('circle', (0.72, 0.68, 0.10), 'orange', 'Bronchopneumonia'), ('circle', (0.30, 0.72, 0.08), 'orange', 'Patchy opacity'), ], 'covid_early': [ ('circle', (0.25, 0.35, 0.10), 'red', 'GGO left'), ('circle', (0.72, 0.32, 0.10), 'red', 'GGO right'), ('arrow', (0.10, 0.22, 0.20, 0.32), 'red', 'Peripheral GGO'), ], 'covid_late': [ ('circle', (0.30, 0.45, 0.15), 'red', 'Consolidation L'), ('circle', (0.68, 0.45, 0.15), 'red', 'Consolidation R'), ('arrow', (0.50, 0.80, 0.42, 0.70), 'yellow', 'Pleural effusion'), ], 'tb_primary': [ ('circle', (0.65, 0.58, 0.09), 'lime', 'Ghon focus'), ('arrow', (0.75, 0.38, 0.67, 0.48), 'lime', 'Hilar lymphadenopathy'), ('circle', (0.68, 0.44, 0.06), 'cyan', 'Hilar LN'), ], 'tb_cavitary': [ ('circle', (0.60, 0.25, 0.12), 'red', 'Cavity'), ('arrow', (0.85, 0.20, 0.70, 0.27), 'red', 'Thick-walled cavity'), ('circle', (0.55, 0.38, 0.08), 'orange', 'Infiltrate'), ], 'tb_miliary': [ ('arrow', (0.15, 0.25, 0.28, 0.35), 'yellow', '1-3mm nodules'), ('arrow', (0.85, 0.30, 0.72, 0.40), 'yellow', 'Uniform distribution'), ('circle', (0.30, 0.50, 0.05), 'yellow', ''), ('circle', (0.55, 0.35, 0.05), 'yellow', ''), ('circle', (0.70, 0.60, 0.05), 'yellow', ''), ], 'ptx_insp': [ ('arrow', (0.12, 0.20, 0.22, 0.30), 'red', 'Visceral pleural line'), ('circle', (0.18, 0.35, 0.08), 'red', 'Avascular zone'), ], 'ptx_exp': [ ('arrow', (0.12, 0.18, 0.20, 0.28), 'red', 'Pleural line'), ('circle', (0.15, 0.40, 0.10), 'red', 'Air space (PTX)'), ('arrow', (0.50, 0.15, 0.50, 0.35), 'cyan', 'No lung markings'), ], 'ptx_skinfold': [ ('arrow', (0.82, 0.25, 0.72, 0.50), 'yellow', 'Skin fold (NOT pleural line)'), ('arrow', (0.78, 0.60, 0.68, 0.68), 'lime', 'Vessels still visible laterally'), ], 'effusion': [ ('circle', (0.72, 0.72, 0.12), 'cyan', 'Effusion'), ('arrow', (0.88, 0.60, 0.78, 0.68), 'cyan', 'Meniscus sign'), ('arrow', (0.88, 0.82, 0.75, 0.78), 'yellow', 'Blunted CP angle'), ], 'edema': [ ('circle', (0.28, 0.45, 0.13), 'red', 'Infiltrate L'), ('circle', (0.65, 0.40, 0.13), 'red', 'Infiltrate R'), ('arrow', (0.05, 0.30, 0.18, 0.38), 'yellow', 'Alveolar oedema'), ], 'measles_pna': [ ('circle', (0.35, 0.40, 0.10), 'orange', 'Perihilar thickening'), ('circle', (0.65, 0.42, 0.10), 'orange', 'Bilateral infiltrate'), ('arrow', (0.50, 0.18, 0.50, 0.32), 'orange', 'Bronchovascular thickening'), ], 'covid_ct': [ ('circle', (0.25, 0.45, 0.12), 'red', 'Consolidation'), ('circle', (0.72, 0.42, 0.12), 'red', 'GGO'), ('arrow', (0.10, 0.25, 0.18, 0.35), 'yellow', 'Crazy paving'), ], } def draw_arrow(draw, x1, y1, x2, y2, color, width=4): draw.line([(x1,y1),(x2,y2)], fill=color, width=width) import math angle = math.atan2(y2-y1, x2-x1) arr_len = 18 arr_angle = 0.45 ax1 = x2 - arr_len*math.cos(angle-arr_angle) ay1 = y2 - arr_len*math.sin(angle-arr_angle) ax2 = x2 - arr_len*math.cos(angle+arr_angle) ay2 = y2 - arr_len*math.sin(angle+arr_angle) draw.polygon([(x2,y2),(ax1,ay1),(ax2,ay2)], fill=color) def annotate(img_path, annots, out_path): img = Image.open(img_path).convert('RGBA') W, H = img.size overlay = Image.new('RGBA', img.size, (0,0,0,0)) draw = ImageDraw.Draw(overlay) color_map = { 'red': (255,50,50,220), 'yellow': (255,230,0,220), 'cyan': (0,220,255,220), 'lime': (50,255,50,220), 'orange': (255,160,0,220), 'white': (255,255,255,220), } try: font = ImageFont.truetype('/usr/share/fonts/truetype/dejavu/DejaVuSans-Bold.ttf', max(14, W//45)) font_sm = ImageFont.truetype('/usr/share/fonts/truetype/dejavu/DejaVuSans-Bold.ttf', max(12, W//55)) except: font = ImageFont.load_default() font_sm = font for ann in annots: t = ann[0]; coords = ann[1]; col_name = ann[2]; label = ann[3] col = color_map.get(col_name, (255,50,50,220)) col_solid = col[:3] + (255,) if t == 'arrow': x1,y1,x2,y2 = int(coords[0]*W), int(coords[1]*H), int(coords[2]*W), int(coords[3]*H) draw_arrow(draw, x1, y1, x2, y2, col, width=max(3, W//160)) if label: # Label at start of arrow bbox = draw.textbbox((0,0), label, font=font_sm) tw = bbox[2]-bbox[0]; th = bbox[3]-bbox[1] tx = max(2, min(x1-tw//2, W-tw-2)) ty = max(2, y1-th-6) draw.rectangle([tx-3, ty-2, tx+tw+3, ty+th+2], fill=(0,0,0,160)) draw.text((tx, ty), label, fill=col_solid, font=font_sm) elif t == 'circle': cx, cy, r = int(coords[0]*W), int(coords[1]*H), int(coords[2]*min(W,H)) lw = max(3, W//140) # Draw circle with thick outline draw.ellipse([cx-r, cy-r, cx+r, cy+r], outline=col, width=lw) # Inner glow draw.ellipse([cx-r-lw, cy-r-lw, cx+r+lw, cy+r+lw], outline=col[:3]+(80,), width=2) if label: bbox = draw.textbbox((0,0), label, font=font_sm) tw = bbox[2]-bbox[0]; th = bbox[3]-bbox[1] tx = max(2, min(cx-tw//2, W-tw-2)) ty = max(2, cy+r+4) draw.rectangle([tx-3, ty-2, tx+tw+3, ty+th+2], fill=(0,0,0,160)) draw.text((tx, ty), label, fill=col_solid, font=font_sm) elif t == 'rect': x1,y1,x2,y2 = int(coords[0]*W),int(coords[1]*H),int(coords[2]*W),int(coords[3]*H) lw = max(3, W//140) draw.rectangle([x1,y1,x2,y2], outline=col, width=lw) composite = Image.alpha_composite(img, overlay).convert('RGB') composite.save(out_path, 'PNG', quality=95) print(f' Annotated: {out_path}') # Fetch and annotate all images for key, (url, title) in IMGS.items(): try: r = requests.get(url, timeout=20) if r.status_code != 200: print(f' FAIL fetch {key}') continue raw_path = f'{OUT_DIR}/{key}_raw.png' with open(raw_path, 'wb') as f: f.write(r.content) annots = ANNOTS.get(key, []) out_path = f'{OUT_DIR}/{key}.png' if annots: annotate(raw_path, annots, out_path) else: import shutil; shutil.copy(raw_path, out_path) os.remove(raw_path) except Exception as e: print(f' ERROR {key}: {e}') print('All images annotated!') "

python3 << 'PYEOF' from reportlab.lib.pagesizes import A4 from reportlab.lib import colors from reportlab.lib.units import cm from reportlab.lib.styles import ParagraphStyle from reportlab.lib.enums import TA_CENTER, TA_LEFT from reportlab.platypus import (SimpleDocTemplate, Paragraph, Spacer, Image, Table, TableStyle, PageBreak, HRFlowable) from reportlab.pdfgen import canvas as rl_canvas import os OUT = "/home/daytona/workspace/cxr-handbook/CXR_Visual_Handbook.pdf" ANN = "/home/daytona/workspace/cxr-handbook/annotated" W, H = A4 # Colors DARK = colors.HexColor("#0D1B2A") BLUE = colors.HexColor("#1B4F8A") LBLUE = colors.HexColor("#D0E8FF") TEAL = colors.HexColor("#0E7C7B") RED = colors.HexColor("#C0392B") LRED = colors.HexColor("#FDECEA") ORANGE = colors.HexColor("#D35400") PURPLE = colors.HexColor("#6C3483") GREEN = colors.HexColor("#1A7A40") GOLD = colors.HexColor("#B7950B") LGOLD = colors.HexColor("#FEF9E7") GREY = colors.HexColor("#F4F6F7") MGREY = colors.HexColor("#566573") WHITE = colors.white def style(name,**kw): return ParagraphStyle(name,**kw) S_title = style("t", fontName="Helvetica-Bold", fontSize=16, textColor=WHITE, leading=20, alignment=TA_CENTER) S_diag = style("d", fontName="Helvetica-Bold", fontSize=13, textColor=WHITE, leading=16, alignment=TA_LEFT) S_sub = style("s", fontName="Helvetica", fontSize=9, textColor=WHITE, leading=12, alignment=TA_LEFT) S_caption = style("ca", fontName="Helvetica-Bold", fontSize=9, textColor=MGREY, leading=12, alignment=TA_CENTER) S_bullet = style("bu", fontName="Helvetica", fontSize=8.5,textColor=colors.HexColor("#2C3E50"), leading=12, leftIndent=12, spaceAfter=2) S_key = style("ke", fontName="Helvetica-Bold", fontSize=8.5,textColor=RED, leading=12, leftIndent=12, spaceAfter=2) def p(t,s=S_caption): return Paragraph(t,s) def bu(t): return Paragraph(f"• {t}", S_bullet) def kp(t): return Paragraph(f"▶ {t}", S_key) def SP(n=0.25): return Spacer(1, n*cm) def HR(c=LBLUE): return HRFlowable(width="100%",thickness=1.5,color=c,spaceAfter=4,spaceBefore=4) def img(fname, w=14*cm, h=10*cm): path = f"{ANN}/{fname}.png" if not os.path.exists(path): return SP(0.1) from PIL import Image as PIL pil = PIL.open(path); iw,ih = pil.size asp = ih/iw if asp*w > h: w = h/asp else: h = asp*w i = Image(path, width=w, height=h); i.hAlign='CENTER' return i def diag_header(title, subtitle, color=BLUE): rows = [[p(f"<b>{title}</b>", S_diag)]] if subtitle: rows.append([p(subtitle, S_sub)]) t = Table(rows, colWidths=[W-4*cm]) t.setStyle(TableStyle([ ('BACKGROUND',(0,0),(-1,-1),color), ('TOPPADDING',(0,0),(-1,-1),8),('BOTTOMPADDING',(0,0),(-1,-1),8), ('LEFTPADDING',(0,0),(-1,-1),14),('RIGHTPADDING',(0,0),(-1,-1),14), ('ROUNDEDCORNERS',[4,4,4,4]) if hasattr(TableStyle,'ROUNDEDCORNERS') else ('BOX',(0,0),(-1,-1),0,color), ])) return t def signs_box(items, bg=LBLUE, border=BLUE): rows = [[i] for i in items] t = Table(rows, colWidths=[W-5*cm]) t.setStyle(TableStyle([ ('BACKGROUND',(0,0),(-1,-1),bg), ('BOX',(0,0),(-1,-1),1.5,border), ('TOPPADDING',(0,0),(-1,-1),4),('BOTTOMPADDING',(0,0),(-1,-1),4), ('LEFTPADDING',(0,0),(-1,-1),10),('RIGHTPADDING',(0,0),(-1,-1),10), ])) return t def two_img(f1, f2, c1="", c2=""): w2 = (W-5*cm)/2 h2 = 7.5*cm i1 = img(f1, w2, h2); i2 = img(f2, w2, h2) data = [[i1,i2],[p(c1),p(c2)]] t = Table(data, colWidths=[w2+0.5*cm, w2+0.5*cm]) t.setStyle(TableStyle([ ('ALIGN',(0,0),(-1,-1),'CENTER'),('VALIGN',(0,0),(-1,-1),'TOP'), ('TOPPADDING',(0,0),(-1,-1),4),('BOTTOMPADDING',(0,0),(-1,-1),4), ])) return t # ── Cover ───────────────────────────────────────────────────────────────────── def cover(c, doc): c.saveState() # Deep dark background c.setFillColor(DARK); c.rect(0,0,W,H,fill=1,stroke=0) # Blue gradient band c.setFillColor(BLUE); c.rect(0, H*0.46, W, H*0.54, fill=1, stroke=0) # Teal accent c.setFillColor(TEAL); c.rect(0, H*0.43, W, 0.035*H, fill=1, stroke=0) # Gold bar c.setFillColor(GOLD); c.rect(0, H*0.405, W, 0.025*H, fill=1, stroke=0) # Title c.setFillColor(WHITE); c.setFont("Helvetica-Bold", 42) c.drawCentredString(W/2, H*0.625, "CHEST X-RAY") c.setFont("Helvetica-Bold", 26) c.drawCentredString(W/2, H*0.565, "VISUAL HANDBOOK") c.setFillColor(GOLD); c.setFont("Helvetica-Bold", 13) c.drawCentredString(W/2, H*0.52, "Pulmonary Department · Intern Physician Edition") c.setFillColor(colors.HexColor("#AACCEE")); c.setFont("Helvetica-Oblique", 11) c.drawCentredString(W/2, H*0.485, "Annotated Images with Arrows & Circles") # Condition list c.setFillColor(WHITE); c.setFont("Helvetica-Bold", 10) conditions = [ "PNEUMONIA · TUBERCULOSIS · PLEURAL EFFUSION", "PNEUMOTHORAX · PULMONARY EDEMA · ARDS", "COPD · LUNG CANCER · INTERSTITIAL LUNG DISEASE", "PULMONARY EMBOLISM · THORACIC TRAUMA", "SARCOIDOSIS · BRONCHIECTASIS · ATELECTASIS", ] y = H*0.355 for c_txt in conditions: c.setFillColor(colors.HexColor("#CCE5FF")) c.drawCentredString(W/2, y, c_txt); y -= 0.042*H # Legend box c.setFillColor(colors.HexColor("#1A1A2E")); c.rect(1.5*cm, 2.8*cm, W-3*cm, 1.6*cm, fill=1, stroke=0) c.setFillColor(colors.HexColor("#FF3232")); c.setFont("Helvetica-Bold", 9) c.drawString(2.2*cm, 4.05*cm, "● Red circle/arrow = pathology location") c.setFillColor(colors.HexColor("#FFE000")) c.drawString(2.2*cm, 3.6*cm, "● Yellow = key radiological sign") c.setFillColor(colors.HexColor("#00DCFF")) c.drawString(W/2+0.5*cm, 4.05*cm, "● Cyan = structural change") c.setFillColor(colors.HexColor("#00FF80")) c.drawString(W/2+0.5*cm, 3.6*cm, "● Green = normal variant/comparison") c.setFillColor(GOLD); c.setFont("Helvetica-Bold", 8) c.drawCentredString(W/2, 1.8*cm, "Murray & Nadel's Respiratory Medicine · Grainger & Allison's Diagnostic Radiology · Goldman-Cecil Medicine") c.setFillColor(WHITE); c.setFont("Helvetica", 8) c.drawCentredString(W/2, 1.2*cm, "All images annotated with pathology markers · June 2026") c.restoreState() def footer(c, doc): c.saveState() c.setFillColor(DARK); c.rect(0,0,W,0.9*cm,fill=1,stroke=0) c.setFillColor(WHITE); c.setFont("Helvetica",7) c.drawCentredString(W/2, 0.3*cm, f"CXR Visual Handbook · Pulmonary Department · Page {doc.page}") c.setFillColor(GOLD); c.setFont("Helvetica-Bold",7) c.drawString(1.5*cm, 0.3*cm, "INTERN EDITION 2026") c.restoreState() # ── Build story ─────────────────────────────────────────────────────────────── story = [PageBreak()] # cover handled by onFirstPage # ══ 1. PNEUMONIA ══════════════════════════════════════════════════════════════ story += [ diag_header("1. LOBAR PNEUMONIA", "Streptococcus pneumoniae · Klebsiella pneumoniae", RED), SP(), img("lobar_pneumonia", 15*cm, 11*cm), SP(0.15), signs_box([ kp("Red circle = lobar consolidation — homogeneous opacity in right upper lobe"), bu("Air bronchogram (red arrow) = air-filled bronchi visible within dense consolidated lung"), bu("Entire lobe opacified · sharp fissure boundary · volume preserved"), bu("Bulging fissure = Klebsiella · Cavitation = Klebsiella/Staph/anaerobes"), ], LRED, RED), PageBreak(), diag_header("2. BRONCHOPNEUMONIA", "Klebsiella · Staphylococcus · Haemophilus influenzae", ORANGE), SP(), img("klebsiella", 15*cm, 11*cm), SP(0.15), signs_box([ kp("Orange circles = patchy multifocal opacities following bronchial tree distribution"), bu("Bilateral lower lobe predominance · inhomogeneous · no lobar boundary"), bu("Nodular centrilobular pattern on CT"), bu("Cavitation = abscess (Klebsiella/Staph)"), ], colors.HexColor("#FEF0E7"), ORANGE), PageBreak(), diag_header("3. ASPIRATION PNEUMONIA", "Right lower lobe · Dependent segment · Anaerobes", colors.HexColor("#7B241C")), SP(), img("aspiration", 15*cm, 11*cm), SP(0.15), signs_box([ kp("Red circle = RLL consolidation with volume loss (dependent segment)"), bu("Cyan arrow = tracheal deviation toward affected side (volume loss)"), bu("Right > left (more vertical right bronchus) · Posterior segments in supine patient"), bu("Complications: lung abscess · empyema · ARDS"), ], LRED, colors.HexColor("#7B241C")), PageBreak(), diag_header("4. ATYPICAL / VIRAL PNEUMONIA", "Mycoplasma · Adenovirus · Influenza · Legionella", PURPLE), SP(), img("atypical_pneumonia", 15*cm, 11*cm), SP(0.15), signs_box([ kp("Yellow circles = bilateral perihilar peribronchial thickening"), bu("Interstitial pattern · reticular infiltrates · NOT lobar"), bu("CXR looks WORSE than patient appears — classic 'walking pneumonia'"), bu("Legionella: rapid progression · pleural effusion 50% · hyponatraemia"), ], colors.HexColor("#F5EEF8"), PURPLE), PageBreak(), diag_header("5. VIRAL PNEUMONIA — MEASLES / PERIHILAR PATTERN", "Bilateral bronchovascular thickening", colors.HexColor("#512E5F")), SP(), img("measles_pna", 15*cm, 11*cm), SP(0.15), signs_box([ kp("Orange circles = patchy bilateral faintly nodular bronchovascular thickening"), bu("Predominantly perihilar distribution · Consistent with viral aetiology"), bu("No lobar consolidation · Interstitial rather than alveolar pattern"), ], colors.HexColor("#F5EEF8"), colors.HexColor("#512E5F")), PageBreak(), ] # ══ COVID-19 ═══════════════════════════════════════════════════════════════════ story += [ diag_header("6. COVID-19 PNEUMONIA — EARLY PHASE", "Bilateral peripheral GGO · Days 1-7", RED), SP(), img("covid_early", 15*cm, 11*cm), SP(0.15), signs_box([ kp("Red circles = bilateral peripheral UPPER LOBE rounded ground-glass opacities"), bu("Red arrow = peripheral location (subpleural) — classic COVID-19"), bu("GGO = hazy opacity with preserved bronchial/vascular markings"), bu("Early exudative phase · Days 1-7 · Often bilateral"), ], LRED, RED), PageBreak(), diag_header("7. COVID-19 PNEUMONIA — ORGANISING PHASE", "Confluent consolidation · Days 7-14", colors.HexColor("#922B21")), SP(), img("covid_late", 15*cm, 11*cm), SP(0.15), signs_box([ kp("Red circles = bilateral multifocal CONSOLIDATION with GGO"), bu("Yellow arrow = small left pleural effusion (blunted CP angle)"), bu("More confluent than early phase · Can progress to ARDS"), bu("CT: crazy paving · reversed halo · traction bronchiectasis"), ], LRED, colors.HexColor("#922B21")), PageBreak(), diag_header("8. COVID-19 / ARDS — CT PATTERN", "Crazy paving · Bilateral consolidation · Fibrosis", colors.HexColor("#641E16")), SP(), img("covid_ct", 15*cm, 11*cm), SP(0.15), signs_box([ kp("Red circles = extensive bilateral consolidation"), kp("GGO (yellow) = ground-glass opacity with interlobular thickening = 'crazy paving'"), bu("Advanced organizing pneumonia / ARDS pattern"), bu("Traction bronchiectasis + fibrosis = irreversible lung injury marker"), ], LRED, colors.HexColor("#641E16")), PageBreak(), ] # ══ TUBERCULOSIS ════════════════════════════════════════════════════════════ story += [ diag_header("9. PRIMARY TUBERCULOSIS", "Ghon Focus + Hilar Lymphadenopathy = Ghon Complex", TEAL), SP(), img("tb_primary", 15*cm, 11*cm), SP(0.15), signs_box([ kp("Lime circle = Ghon focus — right lower lobe consolidation (primary infection site)"), kp("Cyan circle + lime arrow = right hilar lymphadenopathy"), bu("Ghon complex = parenchymal focus + lymphadenopathy"), bu("Middle/lower zone · Any lobe · Cavitation rare in primary TB"), bu("Pleural effusion may occur · Children and young adults"), ], LIGHT_TEAL := colors.HexColor("#D0EFEB"), TEAL), PageBreak(), diag_header("10. CAVITARY (REACTIVATION) TB", "Apical/Posterior Upper Lobe · Thick-walled Cavity", colors.HexColor("#0B5345")), SP(), img("tb_cavitary", 15*cm, 11*cm), SP(0.15), signs_box([ kp("Red circle = thick-walled cavity — RIGHT UPPER LOBE (classic site)"), bu("Red arrow = cavity wall · may contain air-fluid level if secondary infection"), bu("Orange circle = surrounding infiltrate (bronchogenic spread)"), bu("SEND SPUTUM AFB ×3 + GeneXpert immediately"), bu("Most infectious form of TB · Droplet nuclei spread"), ], colors.HexColor("#D0EFEB"), colors.HexColor("#0B5345")), PageBreak(), diag_header("11. MILIARY TUBERCULOSIS", "Haematogenous dissemination · 1-3mm nodules · Bilateral", colors.HexColor("#1A5276")), SP(), img("tb_miliary", 15*cm, 11*cm), SP(0.15), signs_box([ kp("Yellow annotations = innumerable 1-3mm uniform nodules throughout BOTH lungs"), bu("'Millet seed' pattern — randomly distributed, uniform size"), bu("Results from bloodstream dissemination · Profound immunosuppression"), bu("URGENT: screen for TB meningitis (LP) · High mortality if untreated"), bu("Associated with HIV, malnutrition, immunosuppression, children"), ], colors.HexColor("#D6EAF8"), colors.HexColor("#1A5276")), PageBreak(), ] # ══ PLEURAL EFFUSION ═════════════════════════════════════════════════════════ story += [ diag_header("12. PLEURAL EFFUSION", "Meniscus Sign · CP Angle Blunting · Mediastinal Shift", BLUE), SP(), img("effusion", 15*cm, 11*cm), SP(0.15), signs_box([ kp("Cyan circle = pleural fluid — homogeneous basal opacity"), kp("Cyan arrow = MENISCUS SIGN — concave upper border, higher laterally"), bu("Yellow arrow = blunted costophrenic angle"), bu("200-500mL = lateral CP blunting · >1L = up to 4th rib · Massive = contralateral shift"), bu("No shift with large effusion = mesothelioma or ipsilateral collapse"), ], LBLUE, BLUE), PageBreak(), ] # ══ PNEUMOTHORAX ═════════════════════════════════════════════════════════════ story += [ diag_header("13. PNEUMOTHORAX — INSPIRATION vs EXPIRATION", "Visceral Pleural Line · Avascular Zone", ORANGE), SP(), two_img("ptx_insp", "ptx_exp", "Inspiration: visceral pleural line (red arrow) · avascular zone (red circle)", "Expiration: PTX MORE visible (red) · Use expiratory film for small PTX"), SP(0.2), signs_box([ kp("Red arrow = visceral pleural line — thin line separated from chest wall"), bu("Avascular zone (red circle) = no lung markings peripheral to pleural line"), bu("Expiratory film: lung volume decreases, PTX remains same size → MORE apparent"), bu("TENSION PTX: trachea deviates AWAY · complete collapse · treat WITHOUT waiting for CXR"), ], colors.HexColor("#FEF0E7"), ORANGE), PageBreak(), diag_header("14. SKIN FOLD vs PNEUMOTHORAX — IMPORTANT PITFALL", "Do NOT misdiagnose — check for vessels laterally", colors.HexColor("#6E2F00")), SP(), img("ptx_skinfold", 15*cm, 11*cm), SP(0.15), signs_box([ kp("Yellow arrow = SKIN FOLD — NOT a real pneumothorax"), kp("Lime arrow = blood vessels still visible LATERAL to the line → rules out PTX"), bu("Real PTX: thin visceral pleural line · NO vessels lateral to it · stays within chest"), bu("Skin fold: extends beyond chest wall · wide margin · vessels visible throughout"), bu("When in doubt: expiratory film or CT chest"), ], colors.HexColor("#FEF5E7"), colors.HexColor("#6E2F00")), PageBreak(), ] # ══ PULMONARY EDEMA ══════════════════════════════════════════════════════════ story += [ diag_header("15. PULMONARY OEDEMA / HAPE", "Bilateral Alveolar Infiltrates · Before & After Treatment", GREEN), SP(), img("edema", 15*cm, 11*cm), SP(0.15), signs_box([ kp("Red circles = bilateral alveolar infiltrates (oedema)"), bu("Yellow arrow = alveolar oedema pattern — fluffy, bilateral, perihilar"), bu("CARDIOGENIC: bat-wing · cardiomegaly · Kerley B-lines · upper lobe diversion"), bu("NON-CARDIOGENIC (ARDS): normal heart · peripheral · no Kerley lines"), bu("Kerley B-lines = short horizontal lines at lung bases ⊥ to pleura"), ], colors.HexColor("#D5F5E3"), GREEN), PageBreak(), ] # ══ TRAUMA (quick reference page) ══════════════════════════════════════════ story += [ diag_header("16. THORACIC TRAUMA — EMERGENCY CXR FINDINGS", "Recognise immediately · Act without delay", colors.HexColor("#922B21")), SP(0.3), ] trauma_entries = [ ("RIB FRACTURES", "Cortical break/step in rib · Flail chest: ≥2 fractures in ≥3 ribs · Check for underlying PTX/haemothorax"), ("HAEMOTHORAX", "Homogeneous opacity at base · Massive: complete hemithoracic opacification → chest drain (32-36Fr)"), ("PULMONARY CONTUSION", "Patchy consolidation · No lobar boundary · Appears within hours · Worsens at 24-48h · Risk of ARDS"), ("TRAUMATIC PNEUMOTHORAX", "Visceral pleural line · Deep sulcus sign (supine) · Tension → IMMEDIATE needle decompression"), ("AORTIC INJURY", "Widened mediastinum >8cm · Loss of aortic knuckle · Left apical cap · Trachea/NG tube rightward → CT aorta URGENT"), ("DIAPHRAGM RUPTURE", "Elevated hemidiaphragm · Bowel loops in chest · NG tube coiled in chest → surgical repair"), ("CARDIAC TAMPONADE", "'Water-bottle' globular heart · Clear lung fields · Hypotension → emergency pericardiocentesis"), ("PNEUMOMEDIASTINUM", "Air outlining mediastinal structures · Subcutaneous emphysema · Suspect tracheobronchial injury"), ("OESOPHAGEAL RUPTURE", "Pneumomediastinum + left effusion + SC emphysema = Boerhaave → emergency surgery"), ] t_rows = [[ Paragraph("<b><font color='white'>Injury</font></b>", S_bullet), Paragraph("<b><font color='white'>CXR Findings & Action</font></b>", S_bullet) ]] for title, desc in trauma_entries: t_rows.append([ Paragraph(f"<b>{title}</b>", ParagraphStyle("tt", fontName="Helvetica-Bold", fontSize=8.5, textColor=colors.HexColor("#922B21"), leading=12)), Paragraph(desc, ParagraphStyle("td", fontName="Helvetica", fontSize=8.5, textColor=colors.HexColor("#2C3E50"), leading=12)) ]) t_trauma = Table(t_rows, colWidths=[4*cm, W-8*cm]) t_trauma.setStyle(TableStyle([ ('BACKGROUND',(0,0),(-1,0),colors.HexColor("#922B21")), ('ROWBACKGROUNDS',(0,1),(-1,-1),[colors.white, colors.HexColor("#FDECEA")]), ('BOX',(0,0),(-1,-1),1.5,colors.HexColor("#922B21")), ('INNERGRID',(0,0),(-1,-1),0.5,colors.HexColor("#DDDDDD")), ('TOPPADDING',(0,0),(-1,-1),6),('BOTTOMPADDING',(0,0),(-1,-1),6), ('LEFTPADDING',(0,0),(-1,-1),8),('RIGHTPADDING',(0,0),(-1,-1),8), ('VALIGN',(0,0),(-1,-1),'MIDDLE'), ])) story += [t_trauma, PageBreak()] # ══ QUICK REFERENCE — CXR SIGNS ════════════════════════════════════════════ story += [ diag_header("17. CLASSIC CXR SIGNS — QUICK REFERENCE", "Memorise these for ward rounds, emergencies & exams", DARK), SP(0.3), ] signs = [ ("Air bronchogram", "Air-filled bronchi in opacified lung", "Consolidation (pneumonia, ARDS, oedema)"), ("Silhouette sign", "Loss of normal cardiac/diaphragm border", "Localises pathology by lobe"), ("Meniscus sign", "Concave fluid border, higher laterally", "Free pleural effusion"), ("Hampton's Hump", "Peripheral wedge at pleura, flat base", "Pulmonary infarction (PE)"), ("Westermark sign", "Focal oligaemia / avascular zone", "Pulmonary embolism"), ("Bulging fissure", "Fissure bowed away from consolidation", "Klebsiella, severe lobar pneumonia"), ("Deep sulcus sign", "Deep lucent CP angle on supine AP", "Pneumothorax (supine patient)"), ("Bat-wing pattern", "Bilateral perihilar alveolar opacification", "Cardiogenic pulmonary oedema"), ("Eggshell calcification", "Ring calcification of hilar lymph nodes", "Silicosis (pathognomonic)"), ("Water-bottle heart", "Globular enlarged cardiac silhouette", "Pericardial effusion/tamponade"), ("Kerley B-lines", "Short horizontal peripheral basal lines", "Pulm. oedema, lymphangitis carcinomatosa"), ("Golden S sign", "S-shape of right minor fissure", "RUL collapse + central hilar mass"), ("Reversed halo sign", "GGO surrounding central consolidation", "COP, fungal infection, TB"), ("Crazy paving (CT)", "GGO + superimposed interlobular thickening", "COVID-19, PAP, PCP, ARDS"), ("Sail sign", "Triangular thymic opacity in children", "Normal thymus (not consolidation)"), ] s_rows = [[ Paragraph("<b><font color='white'>Sign</font></b>", S_bullet), Paragraph("<b><font color='white'>Appearance</font></b>", S_bullet), Paragraph("<b><font color='white'>Diagnosis</font></b>", S_bullet), ]] for sign, appearance, dx in signs: s_rows.append([ Paragraph(f"<b>{sign}</b>", ParagraphStyle("sr", fontName="Helvetica-Bold", fontSize=8.5, textColor=BLUE, leading=12)), Paragraph(appearance, ParagraphStyle("sa", fontName="Helvetica", fontSize=8.5, textColor=colors.HexColor("#2C3E50"), leading=12)), Paragraph(dx, ParagraphStyle("sd", fontName="Helvetica", fontSize=8.5, textColor=colors.HexColor("#2C3E50"), leading=12)), ]) t_signs = Table(s_rows, colWidths=[4*cm,5.5*cm,4.5*cm]) t_signs.setStyle(TableStyle([ ('BACKGROUND',(0,0),(-1,0),DARK), ('ROWBACKGROUNDS',(0,1),(-1,-1),[colors.white, LBLUE]), ('BOX',(0,0),(-1,-1),1,BLUE), ('INNERGRID',(0,0),(-1,-1),0.5,colors.HexColor("#CCCCCC")), ('TOPPADDING',(0,0),(-1,-1),5),('BOTTOMPADDING',(0,0),(-1,-1),5), ('LEFTPADDING',(0,0),(-1,-1),6),('RIGHTPADDING',(0,0),(-1,-1),6), ('VALIGN',(0,0),(-1,-1),'MIDDLE'), ])) story += [t_signs] # ── Build ───────────────────────────────────────────────────────────────────── doc = SimpleDocTemplate(OUT, pagesize=A4, leftMargin=2*cm, rightMargin=2*cm, topMargin=1.8*cm, bottomMargin=1.8*cm, title="CXR Visual Handbook — Annotated", author="Orris AI — Pulmonary Department") print("Building visual PDF...") doc.build(story, onFirstPage=cover, onLaterPages=footer) import os; sz = os.path.getsize(OUT) print(f"SUCCESS: {OUT}") print(f"File size: {sz/1024:.0f} KB") PYEOF