import os
from reportlab.lib.pagesizes import A4
from reportlab.lib import colors
from reportlab.lib.units import cm, mm
from reportlab.lib.styles import getSampleStyleSheet, ParagraphStyle
from reportlab.lib.enums import TA_LEFT, TA_CENTER, TA_JUSTIFY
from reportlab.platypus import (
    SimpleDocTemplate, Paragraph, Spacer, Image, Table, TableStyle,
    HRFlowable, PageBreak, KeepTogether
)
from reportlab.graphics.shapes import Drawing, Rect, String, Circle, Line, Ellipse, Arrow
from reportlab.graphics import renderPDF
from reportlab.graphics.shapes import Group
from PIL import Image as PILImage
import io

BASE = "/tmp/workspace/wmh-pdf"
OUT  = os.path.join(BASE, "WMH_Brain_MRI_Summary.pdf")

# ── Colour palette ─────────────────────────────────────────────────────────
NAVY      = colors.HexColor("#1a2a4a")
BLUE      = colors.HexColor("#2e5fa3")
LIGHTBLUE = colors.HexColor("#d6e4f7")
TEAL      = colors.HexColor("#1d7874")
ORANGE    = colors.HexColor("#e07b39")
YELLOW    = colors.HexColor("#f5c842")
RED       = colors.HexColor("#c0392b")
WHITE     = colors.white
GRAY      = colors.HexColor("#666666")
LIGHTGRAY = colors.HexColor("#f2f4f7")
DARKGRAY  = colors.HexColor("#333333")

W, H = A4   # 595.27 x 841.89

# ── Styles ──────────────────────────────────────────────────────────────────
styles = getSampleStyleSheet()

def style(name, **kw):
    s = ParagraphStyle(name, **kw)
    return s

H1 = style("H1", fontSize=26, fontName="Helvetica-Bold", textColor=NAVY,
           spaceAfter=6, spaceBefore=4, alignment=TA_CENTER)
H2 = style("H2", fontSize=16, fontName="Helvetica-Bold", textColor=BLUE,
           spaceAfter=4, spaceBefore=12)
H3 = style("H3", fontSize=12, fontName="Helvetica-Bold", textColor=NAVY,
           spaceAfter=3, spaceBefore=8)
BODY = style("BODY", fontSize=10, fontName="Helvetica", textColor=DARKGRAY,
             spaceAfter=4, leading=15, alignment=TA_JUSTIFY)
CAPTION = style("CAPTION", fontSize=8.5, fontName="Helvetica-Oblique",
                textColor=GRAY, spaceAfter=6, alignment=TA_CENTER)
BULLET = style("BULLET", fontSize=10, fontName="Helvetica", textColor=DARKGRAY,
               spaceAfter=3, leading=15, leftIndent=14, bulletIndent=4)
LABEL_ST = style("LABEL", fontSize=8, fontName="Helvetica-Bold",
                 textColor=WHITE, alignment=TA_CENTER)
DISCLAIMER = style("DISCLAIMER", fontSize=8, fontName="Helvetica-Oblique",
                   textColor=GRAY, alignment=TA_CENTER, spaceAfter=4)
SUBTITLE = style("SUBTITLE", fontSize=12, fontName="Helvetica",
                 textColor=BLUE, alignment=TA_CENTER, spaceAfter=6)
FINDING_TITLE = style("FT", fontSize=11, fontName="Helvetica-Bold",
                      textColor=NAVY, spaceAfter=2)

# ── Helper: fit image keeping aspect ratio ──────────────────────────────────
def fit_image(path, max_w, max_h):
    with PILImage.open(path) as im:
        iw, ih = im.size
    ratio = min(max_w / iw, max_h / ih)
    return Image(path, width=iw * ratio, height=ih * ratio)

# ── Helper: coloured badge ──────────────────────────────────────────────────
def badge(text, bg=BLUE, fg=WHITE, w=120, h=18):
    d = Drawing(w, h)
    d.add(Rect(0, 0, w, h, rx=4, ry=4, fillColor=bg, strokeColor=None))
    d.add(String(w/2, 4, text, fontName="Helvetica-Bold", fontSize=9,
                 fillColor=fg, textAnchor="middle"))
    return d

# ── Annotated brain cross-section drawing ───────────────────────────────────
def brain_cross_section_drawing(dw=480, dh=320):
    """Draw a schematic axial brain cross-section with WMH labels."""
    d = Drawing(dw, dh)
    cx, cy = dw / 2, dh / 2 + 10

    # skull outline
    d.add(Ellipse(cx, cy, 200, 160,
                  fillColor=colors.HexColor("#e8e0d0"),
                  strokeColor=colors.HexColor("#9e8c70"), strokeWidth=3))

    # brain parenchyma
    d.add(Ellipse(cx, cy, 185, 148,
                  fillColor=colors.HexColor("#d8cfc0"),
                  strokeColor=colors.HexColor("#b0a090"), strokeWidth=1.5))

    # gray matter rim
    d.add(Ellipse(cx, cy, 185, 148,
                  fillColor=None,
                  strokeColor=colors.HexColor("#c0a080"), strokeWidth=8))

    # white matter zone (inner)
    d.add(Ellipse(cx, cy, 148, 118,
                  fillColor=colors.HexColor("#ede8df"),
                  strokeColor=colors.HexColor("#c8bfb0"), strokeWidth=1))

    # lateral ventricles (left)
    d.add(Ellipse(cx - 42, cy + 8, 28, 14,
                  fillColor=colors.HexColor("#7fb8d4"),
                  strokeColor=colors.HexColor("#5a9ab8"), strokeWidth=1.5))
    # lateral ventricles (right)
    d.add(Ellipse(cx + 42, cy + 8, 28, 14,
                  fillColor=colors.HexColor("#7fb8d4"),
                  strokeColor=colors.HexColor("#5a9ab8"), strokeWidth=1.5))
    # third ventricle
    d.add(Rect(cx - 5, cy - 5, 10, 22,
               fillColor=colors.HexColor("#7fb8d4"),
               strokeColor=colors.HexColor("#5a9ab8"), strokeWidth=1))

    # ── WMH patches (periventricular) ──────────────────────────────────────
    wmh_positions = [
        (cx - 70, cy + 18, 18, 10),  # left periventricular anterior
        (cx - 72, cy - 10, 14, 9),   # left periventricular posterior
        (cx + 52, cy + 18, 18, 10),  # right periventricular anterior
        (cx + 55, cy - 8, 14, 9),    # right periventricular posterior
    ]
    for (ex, ey, ew, eh) in wmh_positions:
        d.add(Ellipse(ex, ey, ew, eh,
                      fillColor=colors.HexColor("#fffacd"),
                      strokeColor=YELLOW, strokeWidth=1.5))

    # ── Deep white matter WMH ─────────────────────────────────────────────
    deep_wmh = [
        (cx - 110, cy + 25, 12, 7),
        (cx + 100, cy - 20, 10, 6),
        (cx - 90,  cy - 30, 10, 6),
    ]
    for (ex, ey, ew, eh) in deep_wmh:
        d.add(Ellipse(ex, ey, ew, eh,
                      fillColor=colors.HexColor("#fff0a0"),
                      strokeColor=ORANGE, strokeWidth=1.2))

    # ── Basal ganglia ────────────────────────────────────────────────────
    d.add(Ellipse(cx - 28, cy - 5, 18, 12,
                  fillColor=colors.HexColor("#b8a898"),
                  strokeColor=colors.HexColor("#908070"), strokeWidth=1))
    d.add(Ellipse(cx + 28, cy - 5, 18, 12,
                  fillColor=colors.HexColor("#b8a898"),
                  strokeColor=colors.HexColor("#908070"), strokeWidth=1))

    # ── Annotations with lines ──────────────────────────────────────────
    anno_font = "Helvetica-Bold"
    anno_small = "Helvetica"

    def annotate(x1, y1, x2, y2, text, sub="", col=NAVY):
        d.add(Line(x1, y1, x2, y2, strokeColor=col, strokeWidth=1.2))
        d.add(Circle(x1, y1, 3, fillColor=col, strokeColor=None))
        d.add(String(x2 + 4, y2 - 2, text, fontName=anno_font,
                     fontSize=8.5, fillColor=col))
        if sub:
            d.add(String(x2 + 4, y2 - 13, sub, fontName=anno_small,
                         fontSize=7.5, fillColor=GRAY))

    # Ventricles label
    annotate(cx - 42, cy + 8, cx - 42, cy + 68, "Lateral Ventricle", "(CSF - dark on FLAIR)", col=colors.HexColor("#2a7aac"))

    # Periventricular WMH label
    annotate(cx - 72, cy + 18, cx - 165, cy + 55,
             "Periventricular WMH", "(bright spots on FLAIR)", col=colors.HexColor("#b8860b"))

    # Deep WMH label
    annotate(cx - 110, cy + 25, cx - 195, cy - 5,
             "Deep White Matter WMH", "(scattered bright foci)", col=ORANGE)

    # White matter label
    annotate(cx, cy - 85, cx + 60, cy - 105,
             "White Matter", "(myelinated axons)", col=GRAY)

    # Cortex label
    annotate(cx + 155, cy + 20, cx + 200, cy + 60,
             "Cortex / Gray Matter", "(normal)", col=colors.HexColor("#6b8e23"))

    # Basal ganglia
    annotate(cx + 28, cy - 5, cx + 160, cy - 35,
             "Basal Ganglia", "(normal)", col=GRAY)

    # Legend boxes at bottom
    lx, ly = 30, 12
    # WMH periventricular
    d.add(Rect(lx, ly, 14, 10, fillColor=colors.HexColor("#fffacd"),
               strokeColor=YELLOW, strokeWidth=1.5))
    d.add(String(lx + 18, ly + 2, "Periventricular WMH",
                 fontName="Helvetica", fontSize=7.5, fillColor=DARKGRAY))
    # WMH deep
    d.add(Rect(lx + 145, ly, 14, 10, fillColor=colors.HexColor("#fff0a0"),
               strokeColor=ORANGE, strokeWidth=1.2))
    d.add(String(lx + 163, ly + 2, "Deep White Matter WMH",
                 fontName="Helvetica", fontSize=7.5, fillColor=DARKGRAY))
    # Ventricle
    d.add(Rect(lx + 320, ly, 14, 10, fillColor=colors.HexColor("#7fb8d4"),
               strokeColor=colors.HexColor("#5a9ab8"), strokeWidth=1.5))
    d.add(String(lx + 338, ly + 2, "CSF / Ventricle",
                 fontName="Helvetica", fontSize=7.5, fillColor=DARKGRAY))

    return d

# ── Fazekas scale drawing ────────────────────────────────────────────────────
def fazekas_drawing(dw=480, dh=110):
    """Illustrate the 3-grade Fazekas scale for WMH severity."""
    d = Drawing(dw, dh)
    labels = ["Grade 0\nNo WMH", "Grade 1\nPunctate\n(Mild)", "Grade 2\nBeginning\nConfluence\n(Moderate)", "Grade 3\nLarge\nConfluent\n(Severe)"]
    wmh_counts = [0, 1, 3, 6]
    wmh_large  = [False, False, True, True]
    box_w = 100
    gap = 20
    start_x = (dw - (4 * box_w + 3 * gap)) / 2

    for i, (label, wcount, wlarge) in enumerate(zip(labels, wmh_counts, wmh_large)):
        bx = start_x + i * (box_w + gap)
        by = 28
        # brain oval
        d.add(Ellipse(bx + box_w/2, by + 36, 40, 32,
                      fillColor=colors.HexColor("#d8cfc0"),
                      strokeColor=colors.HexColor("#b0a090"), strokeWidth=1.5))
        # ventricle
        d.add(Ellipse(bx + box_w/2 - 10, by + 38, 10, 6,
                      fillColor=colors.HexColor("#7fb8d4"),
                      strokeColor=None))
        d.add(Ellipse(bx + box_w/2 + 10, by + 38, 10, 6,
                      fillColor=colors.HexColor("#7fb8d4"),
                      strokeColor=None))
        # draw WMH patches
        wmh_locs = [
            (bx + box_w/2 - 22, by + 42),
            (bx + box_w/2 + 18, by + 42),
            (bx + box_w/2 - 30, by + 28),
            (bx + box_w/2 + 24, by + 28),
            (bx + box_w/2 - 20, by + 28),
            (bx + box_w/2 + 12, by + 55),
        ]
        size = 9 if wlarge else 5
        for j in range(wcount):
            lx2, ly2 = wmh_locs[j]
            d.add(Ellipse(lx2, ly2, size, size * 0.6,
                          fillColor=YELLOW,
                          strokeColor=ORANGE, strokeWidth=0.8))
        # label
        col = [GRAY, colors.HexColor("#8B7000"), ORANGE, RED][i]
        d.add(Rect(bx + 5, 4, box_w - 10, 18, rx=3, ry=3,
                   fillColor=col, strokeColor=None))
        for li, lline in enumerate(label.split("\n")):
            d.add(String(bx + box_w/2, 79 - li * 9, lline,
                         fontName="Helvetica-Bold", fontSize=7.5,
                         fillColor=WHITE, textAnchor="middle"))

    return d

# ── FLAIR vs DWI comparison drawing ─────────────────────────────────────────
def flair_dwi_drawing(dw=480, dh=170):
    d = Drawing(dw, dh)
    cx1 = 120
    cx2 = 360
    cy = 100

    for cx, label, bg, wmh_col, wmh_bright, csf_col, title_col, title in [
        (cx1, "FLAIR", colors.HexColor("#0a1a3a"), colors.HexColor("#fffacd"),
         True, colors.HexColor("#0a1a3a"), colors.HexColor("#5599dd"), "FLAIR Sequence"),
        (cx2, "DWI",   colors.HexColor("#050e1f"), colors.HexColor("#050e1f"),
         False, colors.HexColor("#101828"), colors.HexColor("#44aacc"), "DWI Sequence"),
    ]:
        # background circle (MRI dark background)
        d.add(Ellipse(cx, cy, 88, 72,
                      fillColor=bg, strokeColor=colors.HexColor("#222244"),
                      strokeWidth=2))
        # brain parenchyma
        d.add(Ellipse(cx, cy, 76, 62,
                      fillColor=colors.HexColor("#1e3a5a") if label == "FLAIR" else colors.HexColor("#0d2035"),
                      strokeColor=None))
        # gray matter rim brighter
        d.add(Ellipse(cx, cy, 76, 62,
                      fillColor=None,
                      strokeColor=colors.HexColor("#4a7ab8") if label == "FLAIR" else colors.HexColor("#2a4a6a"),
                      strokeWidth=6))
        # ventricles - dark on FLAIR, dark on DWI
        d.add(Ellipse(cx - 18, cy + 4, 12, 7,
                      fillColor=colors.HexColor("#050e1f"),
                      strokeColor=None))
        d.add(Ellipse(cx + 18, cy + 4, 12, 7,
                      fillColor=colors.HexColor("#050e1f"),
                      strokeColor=None))
        # WMH patches - bright on FLAIR, dark/absent on DWI
        wmh_spots = [(cx - 30, cy + 8), (cx + 22, cy + 8),
                     (cx - 38, cy - 8), (cx + 32, cy - 6)]
        spot_col = colors.HexColor("#fffacd") if wmh_bright else colors.HexColor("#1a2a3a")
        spot_stroke = ORANGE if wmh_bright else colors.HexColor("#1a2a3a")
        for (sx, sy) in wmh_spots:
            d.add(Ellipse(sx, sy, 8, 5,
                          fillColor=spot_col,
                          strokeColor=spot_stroke, strokeWidth=0.8))

        # title bar
        d.add(Rect(cx - 80, cy + 82, 160, 20, rx=4, ry=4,
                   fillColor=title_col, strokeColor=None))
        d.add(String(cx, cy + 88, title,
                     fontName="Helvetica-Bold", fontSize=10,
                     fillColor=WHITE, textAnchor="middle"))

    # Arrow between them
    d.add(Line(210, cy, 270, cy, strokeColor=BLUE, strokeWidth=2))
    d.add(String(240, cy + 6, "compare", fontName="Helvetica", fontSize=8,
                 fillColor=BLUE, textAnchor="middle"))

    # Labels below FLAIR
    d.add(String(cx1, cy - 90, "WMH = BRIGHT (chronic)",
                 fontName="Helvetica-Bold", fontSize=8.5,
                 fillColor=colors.HexColor("#c8a800"), textAnchor="middle"))
    d.add(String(cx1, cy - 102, "Ventricles = DARK (suppressed)",
                 fontName="Helvetica", fontSize=7.5,
                 fillColor=colors.HexColor("#8899aa"), textAnchor="middle"))

    # Labels below DWI
    d.add(String(cx2, cy - 90, "No acute infarct (no bright spots)",
                 fontName="Helvetica-Bold", fontSize=8.5,
                 fillColor=colors.HexColor("#44cc88"), textAnchor="middle"))
    d.add(String(cx2, cy - 102, "Normal = all dark/uniform",
                 fontName="Helvetica", fontSize=7.5,
                 fillColor=colors.HexColor("#8899aa"), textAnchor="middle"))

    return d


# ── Build document ───────────────────────────────────────────────────────────
doc = SimpleDocTemplate(
    OUT,
    pagesize=A4,
    leftMargin=2*cm, rightMargin=2*cm,
    topMargin=2.2*cm, bottomMargin=2*cm,
    title="Brain MRI – White Matter Hyperintensities Summary",
    author="Orris Medical AI"
)

story = []

# ═══════════════════════════════════════════════════════════════════════
# PAGE 1 – COVER
# ═══════════════════════════════════════════════════════════════════════

# Top colour bar
story.append(HRFlowable(width="100%", thickness=8, color=NAVY, spaceAfter=10))

story.append(Paragraph("Brain MRI Analysis Report", H1))
story.append(Paragraph("White Matter Hyperintensities (WMH) – Location & Clinical Significance", SUBTITLE))
story.append(HRFlowable(width="100%", thickness=2, color=BLUE, spaceAfter=16))

# Key info box
info_data = [
    [Paragraph("<b>Imaging Modality</b>", BODY), Paragraph("MRI Brain – FLAIR + DWI sequences", BODY)],
    [Paragraph("<b>Primary Finding</b>", BODY), Paragraph("Mild periventricular & deep white matter hyperintensities", BODY)],
    [Paragraph("<b>Clinical Significance</b>", BODY), Paragraph("Chronic small vessel ischemic disease (leukoaraiosis)", BODY)],
    [Paragraph("<b>Acute Finding</b>", BODY), Paragraph("None – No restricted diffusion on DWI (no acute stroke)", BODY)],
    [Paragraph("<b>Report Date</b>", BODY), Paragraph("July 15, 2026", BODY)],
]
info_table = Table(info_data, colWidths=[5*cm, 11*cm])
info_table.setStyle(TableStyle([
    ("BACKGROUND", (0, 0), (0, -1), LIGHTBLUE),
    ("BACKGROUND", (1, 0), (1, -1), colors.HexColor("#f8faff")),
    ("TEXTCOLOR", (0, 0), (-1, -1), DARKGRAY),
    ("FONTNAME", (0, 0), (-1, -1), "Helvetica"),
    ("FONTSIZE", (0, 0), (-1, -1), 10),
    ("GRID", (0, 0), (-1, -1), 0.5, colors.HexColor("#c0ccdd")),
    ("ROWBACKGROUNDS", (0, 0), (-1, -1), [LIGHTBLUE, colors.HexColor("#f8faff")]),
    ("TOPPADDING", (0, 0), (-1, -1), 7),
    ("BOTTOMPADDING", (0, 0), (-1, -1), 7),
    ("LEFTPADDING", (0, 0), (-1, -1), 10),
    ("ROUNDEDCORNERS", [4]),
]))
story.append(info_table)
story.append(Spacer(1, 0.5*cm))

# Patient MRI thumbnail on cover
story.append(Paragraph("Patient MRI Film", H3))
story.append(fit_image("/tmp/workspace/wmh-pdf/images/patient_mri.jpg", 14*cm, 11*cm))
story.append(Paragraph(
    "Figure 1. Original patient MRI film (MEDVISION). Top three rows: FLAIR sequence showing "
    "white matter hyperintensities. Bottom three rows: DWI sequence with no evidence of acute infarction.",
    CAPTION
))

story.append(HRFlowable(width="100%", thickness=1, color=LIGHTBLUE, spaceBefore=8, spaceAfter=6))
story.append(Paragraph(
    "DISCLAIMER: This report is generated by Orris AI for educational purposes only. It does not constitute "
    "a formal radiological diagnosis. All clinical decisions must be made by a qualified physician reviewing "
    "original DICOM images alongside patient history.",
    DISCLAIMER
))

story.append(PageBreak())

# ═══════════════════════════════════════════════════════════════════════
# PAGE 2 – WHAT ARE WMH? + LABELED DIAGRAM
# ═══════════════════════════════════════════════════════════════════════

story.append(HRFlowable(width="100%", thickness=5, color=NAVY, spaceAfter=8))
story.append(Paragraph("1. What Are White Matter Hyperintensities?", H2))

story.append(Paragraph(
    "White Matter Hyperintensities (WMH), also called <b>leukoaraiosis</b>, are areas of the brain "
    "that appear abnormally bright (hyperintense) on FLAIR and T2-weighted MRI sequences. They represent "
    "damage to the white matter – the deep brain tissue made up of myelinated nerve fibers (axons) that "
    "connect different brain regions.", BODY))

story.append(Paragraph(
    "WMH are extremely common in older adults and in patients with cardiovascular risk factors. "
    "They are a hallmark of <b>cerebral small vessel disease (SVD)</b> – chronic damage to the "
    "tiny arteries (arterioles) that supply the deep brain white matter.", BODY))

# Two-column bullet table
bullet_data = [
    [Paragraph("<b>Location in the Brain</b>", H3), Paragraph("<b>What They Represent</b>", H3)],
    [
        Paragraph(
            "• <b>Periventricular</b> – immediately surrounding the lateral ventricles<br/>"
            "• <b>Deep white matter</b> – frontal, parietal, temporal lobes<br/>"
            "• <b>Subcortical</b> – just beneath the cortex<br/>"
            "• <b>Infratentorial</b> – brainstem, cerebellum (less common)", BULLET),
        Paragraph(
            "• Demyelination (loss of myelin sheath around axons)<br/>"
            "• Axonal loss and gliosis (scar tissue formation)<br/>"
            "• Enlarged perivascular spaces<br/>"
            "• Chronic ischemia from small artery thickening", BULLET)
    ],
]
btable = Table(bullet_data, colWidths=[8*cm, 8*cm])
btable.setStyle(TableStyle([
    ("BACKGROUND", (0, 0), (-1, 0), NAVY),
    ("TEXTCOLOR", (0, 0), (-1, 0), WHITE),
    ("BACKGROUND", (0, 1), (0, 1), colors.HexColor("#eef4ff")),
    ("BACKGROUND", (1, 1), (1, 1), colors.HexColor("#fff8ee")),
    ("GRID", (0, 0), (-1, -1), 0.5, colors.HexColor("#c0ccdd")),
    ("TOPPADDING", (0, 0), (-1, -1), 8),
    ("BOTTOMPADDING", (0, 0), (-1, -1), 8),
    ("LEFTPADDING", (0, 0), (-1, -1), 8),
    ("VALIGN", (0, 0), (-1, -1), "TOP"),
]))
story.append(btable)
story.append(Spacer(1, 0.4*cm))

story.append(Paragraph("Labeled Brain Cross-Section – WMH Locations", H3))
story.append(brain_cross_section_drawing(480, 320))
story.append(Paragraph(
    "Figure 2. Schematic axial cross-section of the brain showing the two main locations of WMH: "
    "periventricular (immediately adjacent to the lateral ventricles, shown in pale yellow) and "
    "deep white matter (scattered foci, shown in orange-yellow). The lateral ventricles appear "
    "blue (CSF). Basal ganglia are shown in gray.",
    CAPTION
))

story.append(PageBreak())

# ═══════════════════════════════════════════════════════════════════════
# PAGE 3 – FLAIR vs DWI + FAZEKAS SCALE
# ═══════════════════════════════════════════════════════════════════════

story.append(HRFlowable(width="100%", thickness=5, color=NAVY, spaceAfter=8))
story.append(Paragraph("2. How WMH Look on MRI – FLAIR vs DWI", H2))

story.append(Paragraph(
    "Two key sequences are used to characterise WMH and distinguish them from acute pathology:", BODY))

story.append(flair_dwi_drawing(480, 170))
story.append(Paragraph(
    "Figure 3. Schematic comparison of FLAIR (left) and DWI (right) appearances. "
    "On FLAIR, WMH appear as bright (hyperintense) foci; ventricles are dark. "
    "On DWI, the same WMH are dark – absence of bright signal confirms no acute stroke.",
    CAPTION
))
story.append(Spacer(1, 0.3*cm))

seq_data = [
    [Paragraph("<b>Sequence</b>", LABEL_ST), Paragraph("<b>WMH Appearance</b>", LABEL_ST),
     Paragraph("<b>Why It Matters</b>", LABEL_ST)],
    [Paragraph("FLAIR", BODY), Paragraph("Bright / hyperintense", BODY),
     Paragraph("Best for detecting WMH, periventricular halos, subcortical lesions", BODY)],
    [Paragraph("DWI", BODY), Paragraph("Dark / isointense", BODY),
     Paragraph("Normal DWI confirms lesions are CHRONIC, not an acute stroke", BODY)],
    [Paragraph("T1", BODY), Paragraph("Dark / hypointense", BODY),
     Paragraph("Old lacunar infarcts appear as black holes on T1", BODY)],
    [Paragraph("T2", BODY), Paragraph("Bright / hyperintense", BODY),
     Paragraph("Shows same changes as FLAIR but with CSF also bright (less contrast)", BODY)],
]
seq_table = Table(seq_data, colWidths=[3*cm, 4.5*cm, 8.5*cm])
seq_table.setStyle(TableStyle([
    ("BACKGROUND", (0, 0), (-1, 0), NAVY),
    ("TEXTCOLOR", (0, 0), (-1, 0), WHITE),
    ("ROWBACKGROUNDS", (0, 1), (-1, -1), [colors.HexColor("#f8faff"), WHITE]),
    ("GRID", (0, 0), (-1, -1), 0.5, colors.HexColor("#c0ccdd")),
    ("TOPPADDING", (0, 0), (-1, -1), 7),
    ("BOTTOMPADDING", (0, 0), (-1, -1), 7),
    ("LEFTPADDING", (0, 0), (-1, -1), 8),
    ("FONTNAME", (0, 0), (-1, 0), "Helvetica-Bold"),
    ("FONTSIZE", (0, 0), (-1, -1), 9.5),
    ("VALIGN", (0, 0), (-1, -1), "MIDDLE"),
]))
story.append(seq_table)
story.append(Spacer(1, 0.5*cm))

story.append(Paragraph("3. Grading WMH Severity – Fazekas Scale", H2))
story.append(Paragraph(
    "The <b>Fazekas scale</b> is the most widely used system to grade WMH severity on MRI. "
    "This patient's scan shows Grade 1-2 changes (mild-to-moderate punctate/beginning confluence).", BODY))

story.append(fazekas_drawing(480, 110))
story.append(Paragraph(
    "Figure 4. Schematic Fazekas grading scale for WMH severity. Yellow foci = WMH patches. "
    "Grade 0 = normal; Grade 1 = punctate foci (mild); Grade 2 = beginning confluence (moderate); "
    "Grade 3 = large confluent areas (severe). This patient's scan corresponds to Grade 1-2.",
    CAPTION
))

story.append(Spacer(1, 0.4*cm))

# Reference diagram from Radiopaedia
story.append(Paragraph("Reference: Small Vessel Disease Brain Anatomy", H3))
story.append(fit_image("/tmp/workspace/wmh-pdf/images/svd_diagram.jpg", 10*cm, 8*cm))
story.append(Paragraph(
    "Figure 5. Anatomical diagram (Radiopaedia) illustrating features of cerebral small vessel disease "
    "including WMH location, lacunes, microbleeds, and perivascular spaces in an axial brain cross-section. "
    "This patient's scan shows WMH only (no lacunes, no microbleeds identified on available sequences).",
    CAPTION
))

story.append(PageBreak())

# ═══════════════════════════════════════════════════════════════════════
# PAGE 4 – CAUSES, CLINICAL SIGNIFICANCE, MANAGEMENT
# ═══════════════════════════════════════════════════════════════════════

story.append(HRFlowable(width="100%", thickness=5, color=NAVY, spaceAfter=8))
story.append(Paragraph("4. Causes and Risk Factors", H2))

causes_data = [
    [Paragraph("<b>Risk Factor</b>", LABEL_ST),
     Paragraph("<b>Mechanism</b>", LABEL_ST),
     Paragraph("<b>Effect on White Matter</b>", LABEL_ST)],
    [Paragraph("Hypertension", BODY),
     Paragraph("Arteriolosclerosis – thickening/stiffening of small artery walls", BODY),
     Paragraph("Reduced perfusion to deep white matter watershed zones", BODY)],
    [Paragraph("Diabetes mellitus", BODY),
     Paragraph("Endothelial dysfunction; glycation of vessel walls", BODY),
     Paragraph("Microangiopathy; accelerated small vessel damage", BODY)],
    [Paragraph("Age (>60 years)", BODY),
     Paragraph("Natural stiffening of arteries (arteriosclerosis)", BODY),
     Paragraph("Progressive ischemic white matter rarefaction", BODY)],
    [Paragraph("Hyperlipidaemia", BODY),
     Paragraph("Lipid deposits in arterial walls", BODY),
     Paragraph("Promotes atheromatous change in small vessels", BODY)],
    [Paragraph("Smoking", BODY),
     Paragraph("Oxidative stress; vasospasm; accelerated atherosclerosis", BODY),
     Paragraph("Worsens all above mechanisms", BODY)],
    [Paragraph("Atrial fibrillation", BODY),
     Paragraph("Microemboli from cardiac source", BODY),
     Paragraph("Small embolic infarcts appear as WMH", BODY)],
]
causes_table = Table(causes_data, colWidths=[3.5*cm, 6.5*cm, 6*cm])
causes_table.setStyle(TableStyle([
    ("BACKGROUND", (0, 0), (-1, 0), TEAL),
    ("TEXTCOLOR", (0, 0), (-1, 0), WHITE),
    ("ROWBACKGROUNDS", (0, 1), (-1, -1), [colors.HexColor("#f0faf8"), WHITE]),
    ("GRID", (0, 0), (-1, -1), 0.5, colors.HexColor("#aacccc")),
    ("TOPPADDING", (0, 0), (-1, -1), 7),
    ("BOTTOMPADDING", (0, 0), (-1, -1), 7),
    ("LEFTPADDING", (0, 0), (-1, -1), 8),
    ("FONTNAME", (0, 0), (-1, 0), "Helvetica-Bold"),
    ("FONTSIZE", (0, 0), (-1, -1), 9),
    ("VALIGN", (0, 0), (-1, -1), "TOP"),
]))
story.append(causes_table)
story.append(Spacer(1, 0.5*cm))

story.append(Paragraph("5. Clinical Significance", H2))
story.append(Paragraph(
    "WMH are not merely incidental findings – they carry important clinical implications, "
    "especially when moderate to severe:", BODY))

sig_items = [
    ("<b>Cognitive impairment:</b>", "WMH disrupt white matter tracts connecting frontal-subcortical circuits. "
     "Moderate-severe WMH are associated with slowed processing speed, executive dysfunction, and vascular dementia."),
    ("<b>Stroke risk:</b>", "WMH independently increase future stroke risk (both ischemic and hemorrhagic). "
     "Presence of WMH roughly doubles stroke risk."),
    ("<b>Gait disturbance:</b>", "Periventricular WMH affecting motor tracts can cause small-stepped gait and falls."),
    ("<b>Mood disorders:</b>", "Deep white matter lesions are associated with late-onset depression ('vascular depression')."),
    ("<b>Bladder dysfunction:</b>", "Frontal WMH can cause urinary urgency/incontinence by interrupting inhibitory pathways."),
]
for title_text, body_text in sig_items:
    story.append(Paragraph(f"{title_text} {body_text}", BULLET))

story.append(Spacer(1, 0.4*cm))
story.append(Paragraph("6. Management Recommendations", H2))

mgmt_data = [
    [Paragraph("<b>Goal</b>", LABEL_ST), Paragraph("<b>Action</b>", LABEL_ST)],
    [Paragraph("Control hypertension", BODY),
     Paragraph("Target BP <130/80 mmHg; first-line agents: ACE inhibitor or ARB", BODY)],
    [Paragraph("Manage diabetes", BODY),
     Paragraph("HbA1c <7%; GLP-1 agonists have additional vascular benefit", BODY)],
    [Paragraph("Statin therapy", BODY),
     Paragraph("If dyslipidaemia present; LDL-C <1.8 mmol/L in high-risk patients", BODY)],
    [Paragraph("Antiplatelet therapy", BODY),
     Paragraph("Aspirin 75-100 mg/day if cerebrovascular disease confirmed", BODY)],
    [Paragraph("Lifestyle modification", BODY),
     Paragraph("Smoking cessation, aerobic exercise, Mediterranean diet", BODY)],
    [Paragraph("Cognitive monitoring", BODY),
     Paragraph("Annual cognitive screening (MoCA); neuropsychology if symptomatic", BODY)],
    [Paragraph("Follow-up MRI", BODY),
     Paragraph("Repeat MRI in 2-3 years to monitor WMH progression", BODY)],
]
mgmt_table = Table(mgmt_data, colWidths=[4.5*cm, 11.5*cm])
mgmt_table.setStyle(TableStyle([
    ("BACKGROUND", (0, 0), (-1, 0), NAVY),
    ("TEXTCOLOR", (0, 0), (-1, 0), WHITE),
    ("ROWBACKGROUNDS", (0, 1), (-1, -1), [colors.HexColor("#f8faff"), WHITE]),
    ("GRID", (0, 0), (-1, -1), 0.5, colors.HexColor("#c0ccdd")),
    ("TOPPADDING", (0, 0), (-1, -1), 7),
    ("BOTTOMPADDING", (0, 0), (-1, -1), 7),
    ("LEFTPADDING", (0, 0), (-1, -1), 8),
    ("FONTNAME", (0, 0), (-1, 0), "Helvetica-Bold"),
    ("FONTSIZE", (0, 0), (-1, -1), 9.5),
    ("VALIGN", (0, 0), (-1, -1), "TOP"),
]))
story.append(mgmt_table)

story.append(Spacer(1, 0.5*cm))
story.append(HRFlowable(width="100%", thickness=1, color=LIGHTBLUE, spaceAfter=6))

# WMH progression diagram
story.append(Paragraph("7. How WMH Progress Over Time", H2))
story.append(fit_image("/tmp/workspace/wmh-pdf/images/wmh_progression.jpg", 13*cm, 9*cm))
story.append(Paragraph(
    "Figure 6. Dynamic progression model of WMH and small vessel disease (Wardlaw et al., JAH 2015). "
    "Periventricular and deep WMH (pale yellow) can increase, occasionally shrink, and lead to atrophy "
    "of overlying cortex (blue). Acute small subcortical infarcts (white) may cavitate, persist as WMH caps, "
    "or cause downstream white matter tract damage.",
    CAPTION
))

story.append(PageBreak())

# ═══════════════════════════════════════════════════════════════════════
# PAGE 5 – SUMMARY FINDINGS TABLE + KEY MESSAGES
# ═══════════════════════════════════════════════════════════════════════

story.append(HRFlowable(width="100%", thickness=5, color=NAVY, spaceAfter=8))
story.append(Paragraph("8. Summary of MRI Findings – This Patient", H2))

summary_data = [
    [Paragraph("<b>Brain Region</b>", LABEL_ST),
     Paragraph("<b>Sequence</b>", LABEL_ST),
     Paragraph("<b>Finding</b>", LABEL_ST),
     Paragraph("<b>Significance</b>", LABEL_ST)],
    [Paragraph("Periventricular white matter", BODY),
     Paragraph("FLAIR", BODY),
     Paragraph("Small bright hyperintensities adjacent to lateral ventricles", BODY),
     Paragraph("Chronic small vessel disease", BODY)],
    [Paragraph("Deep white matter\n(frontal/parietal)", BODY),
     Paragraph("FLAIR", BODY),
     Paragraph("Scattered mild hyperintense foci", BODY),
     Paragraph("Chronic microvascular ischemia", BODY)],
    [Paragraph("Lateral ventricles", BODY),
     Paragraph("FLAIR", BODY),
     Paragraph("Normal size, symmetric; CSF correctly dark", BODY),
     Paragraph("No hydrocephalus", BODY)],
    [Paragraph("Cerebral hemispheres", BODY),
     Paragraph("DWI", BODY),
     Paragraph("No restricted diffusion (no bright areas)", BODY),
     Paragraph("No acute ischemic stroke", BODY)],
    [Paragraph("Brainstem/cerebellum", BODY),
     Paragraph("FLAIR + DWI", BODY),
     Paragraph("Structurally intact; no focal lesion", BODY),
     Paragraph("Normal posterior fossa", BODY)],
    [Paragraph("Cortex / sulci", BODY),
     Paragraph("FLAIR", BODY),
     Paragraph("Mild sulcal widening (age-appropriate)", BODY),
     Paragraph("Mild diffuse cortical atrophy", BODY)],
    [Paragraph("Basal ganglia / thalami", BODY),
     Paragraph("FLAIR", BODY),
     Paragraph("No focal signal change", BODY),
     Paragraph("Normal deep grey matter", BODY)],
    [Paragraph("Midline structures", BODY),
     Paragraph("FLAIR", BODY),
     Paragraph("No shift; corpus callosum intact", BODY),
     Paragraph("No mass effect", BODY)],
]
summary_table = Table(summary_data, colWidths=[4*cm, 2.5*cm, 6*cm, 3.5*cm])
summary_table.setStyle(TableStyle([
    ("BACKGROUND", (0, 0), (-1, 0), NAVY),
    ("TEXTCOLOR", (0, 0), (-1, 0), WHITE),
    ("ROWBACKGROUNDS", (0, 1), (-1, -1), [colors.HexColor("#fff9e6"), WHITE]),
    ("GRID", (0, 0), (-1, -1), 0.5, colors.HexColor("#c0ccdd")),
    ("TOPPADDING", (0, 0), (-1, -1), 6),
    ("BOTTOMPADDING", (0, 0), (-1, -1), 6),
    ("LEFTPADDING", (0, 0), (-1, -1), 6),
    ("FONTNAME", (0, 0), (-1, 0), "Helvetica-Bold"),
    ("FONTSIZE", (0, 0), (-1, -1), 8.5),
    ("VALIGN", (0, 0), (-1, -1), "TOP"),
]))
story.append(summary_table)
story.append(Spacer(1, 0.5*cm))

story.append(Paragraph("Key Clinical Messages", H2))
key_msgs = [
    ("No emergency finding.", "There is no acute stroke, hemorrhage, tumor, or midline shift on this scan."),
    ("Chronic vascular change.", "The WMH represent cumulative damage from long-standing small vessel disease – "
     "not a new event."),
    ("Modifiable risk factors.", "Most causes of WMH (hypertension, diabetes, smoking) are treatable. "
     "Aggressive risk factor control can slow progression."),
    ("Cognitive vigilance.", "Even mild WMH warrant monitoring for early cognitive or gait changes."),
    ("Follow-up imaging.", "A repeat MRI in 2-3 years is recommended to track WMH evolution "
     "and ensure no new lesions develop."),
]
for num, (msg_title, msg_body) in enumerate(key_msgs, 1):
    d = Drawing(480, 38)
    d.add(Rect(0, 4, 480, 32, rx=5, ry=5,
               fillColor=LIGHTBLUE if num % 2 == 0 else colors.HexColor("#fff8e8"),
               strokeColor=BLUE, strokeWidth=0.5))
    d.add(Rect(0, 4, 8, 32, rx=0, ry=0,
               fillColor=BLUE if num % 2 == 0 else ORANGE, strokeColor=None))
    d.add(String(20, 24, f"{num}. {msg_title}", fontName="Helvetica-Bold",
                 fontSize=9.5, fillColor=NAVY))
    d.add(String(20, 11, msg_body, fontName="Helvetica",
                 fontSize=8.5, fillColor=DARKGRAY))
    story.append(d)
    story.append(Spacer(1, 2))

story.append(Spacer(1, 0.6*cm))
story.append(HRFlowable(width="100%", thickness=2, color=NAVY, spaceAfter=8))
story.append(Paragraph(
    "References: Wardlaw JM et al. (2013) Neuroimaging standards for research into small vessel disease. "
    "Lancet Neurol 12(8):822-38. | Fazekas F et al. (1987) CT and MRI rating of white matter lesions. "
    "AJNR 8(2):421-6. | Hachinski V et al. (1987) Leukoaraiosis. Arch Neurol 44(1):21-3.",
    DISCLAIMER
))
story.append(Paragraph(
    "DISCLAIMER: This document is generated by Orris AI for educational/informational purposes only. "
    "It is NOT a substitute for formal radiological reporting by a qualified radiologist. "
    "Clinical decisions must be based on complete clinical assessment.",
    DISCLAIMER
))

doc.build(story)
print(f"PDF saved to: {OUT}")

Successfully replaced 1 instance(s) of the string in '/tmp/workspace/wmh-pdf/build_pdf.py'