cd /tmp/workspace/attachments/e0bd04dc-1d8c-42c7-92cc-2876d15d13b8/ && python3 -c " from docx import Document doc = Document('DIAGNOSTIC TOOL FOR PEDIATRIC RESPIRATORY DISORDERS – PROPOSAL OF AN AI BASED MODEL.docx') for i, para in enumerate(doc.paragraphs): if para.text.strip(): print(f'[{i}] STYLE={para.style.name!r} | {para.text}') "

exhaled breath temperature airway inflammation biomarker

artificial intelligence deep learning pediatric respiratory diagnosis

Kashyapa Samhita Vedanadhyaya Swasa Nishtanatyurasa Atyushnam pediatric respiratory Ayurveda reference

AI machine learning respiratory sound cough asthma diagnosis children

machine learning respiratory sound classification asthma COPD

exhaled breath temperature asthma children measurement

Kashyapa Samhita Swasa pediatric Kuchla Nishtanatyurasa exhaled hot breath reference chapter

mkdir -p /tmp/workspace/review-article && python3 << 'PYEOF' from docx import Document from docx.shared import Pt, Inches, RGBColor from docx.enum.text import WD_ALIGN_PARAGRAPH from docx.oxml.ns import qn from docx.oxml import OxmlElement import copy doc = Document() # --- Page margins --- from docx.shared import Inches section = doc.sections[0] section.top_margin = Inches(1) section.bottom_margin = Inches(1) section.left_margin = Inches(1.25) section.right_margin = Inches(1.25) # --- Default Normal style --- normal_style = doc.styles['Normal'] normal_style.font.name = 'Times New Roman' normal_style.font.size = Pt(12) # ---- Heading helpers ---- def add_heading(doc, text, level=1, color=None, center=False): p = doc.add_paragraph() run = p.add_run(text) run.bold = True if level == 1: run.font.size = Pt(16) elif level == 2: run.font.size = Pt(14) elif level == 3: run.font.size = Pt(12) run.underline = True else: run.font.size = Pt(12) if color: run.font.color.rgb = RGBColor(*color) if center: p.alignment = WD_ALIGN_PARAGRAPH.CENTER p.paragraph_format.space_before = Pt(12) p.paragraph_format.space_after = Pt(6) return p def add_body(doc, text, indent=False, bold=False): p = doc.add_paragraph() run = p.add_run(text) run.font.name = 'Times New Roman' run.font.size = Pt(12) run.bold = bold p.alignment = WD_ALIGN_PARAGRAPH.JUSTIFY p.paragraph_format.space_after = Pt(6) if indent: p.paragraph_format.left_indent = Inches(0.3) return p def add_bullet(doc, text, indent=0.3): p = doc.add_paragraph(style='List Bullet') run = p.add_run(text) run.font.name = 'Times New Roman' run.font.size = Pt(12) p.paragraph_format.left_indent = Inches(indent) p.paragraph_format.space_after = Pt(4) return p def add_sub_bullet(doc, text): return add_bullet(doc, text, indent=0.5) def add_space(doc): p = doc.add_paragraph() p.paragraph_format.space_after = Pt(2) # ============================================================ # TITLE PAGE # ============================================================ p = doc.add_paragraph() p.alignment = WD_ALIGN_PARAGRAPH.CENTER r = p.add_run("DIAGNOSTIC TOOL FOR PEDIATRIC RESPIRATORY DISORDERS") r.bold = True r.font.size = Pt(18) r.font.color.rgb = RGBColor(0x1A, 0x37, 0x7A) p2 = doc.add_paragraph() p2.alignment = WD_ALIGN_PARAGRAPH.CENTER r2 = p2.add_run("PROPOSAL OF AN AI-BASED MODEL") r2.bold = True r2.font.size = Pt(16) r2.font.color.rgb = RGBColor(0x1A, 0x37, 0x7A) add_space(doc) p3 = doc.add_paragraph() p3.alignment = WD_ALIGN_PARAGRAPH.CENTER r3 = p3.add_run("A Review Article") r3.italic = True r3.font.size = Pt(13) r3.font.color.rgb = RGBColor(0x55, 0x55, 0x55) add_space(doc) # Type label p4 = doc.add_paragraph() p4.alignment = WD_ALIGN_PARAGRAPH.CENTER r4 = p4.add_run("Review Article | Integrative Pediatric Respiratory Medicine") r4.font.size = Pt(11) r4.italic = True doc.add_page_break() # ============================================================ # ABSTRACT # ============================================================ add_heading(doc, "ABSTRACT", level=1, color=(0x1A,0x37,0x7A), center=True) add_body(doc, "Respiratory disorders represent a leading cause of pediatric morbidity and mortality worldwide. " "Conventional diagnostic methods, though effective, present significant challenges in children due " "to their invasive nature, radiation exposure, and dependence on patient cooperation. Expiratory Air " "Temperature (EAT), described clinically as a non-invasive physiological parameter, reflects airway " "inflammation, mucosal blood flow, and metabolic activity of the respiratory tract. Interestingly, " "the classical Ayurvedic text Kashyapa Samhita, in its chapter Vedanadhyaya, describes " "Nishtanatyurasa Atyushnam (exhaling hot breath) as one of the cardinal symptoms of Swasa " "(respiratory disease) in children, providing an ancient conceptual basis for EAT assessment. " "This review proposes a novel AI-based diagnostic model that integrates EAT measurements, " "environmental contextual data, and basic clinical parameters to support early detection and " "monitoring of pediatric respiratory disorders. The model employs machine learning and deep learning " "techniques trained on EAT pattern profiles from healthy children and those with various respiratory " "conditions. By bridging classical Ayurvedic clinical observation with contemporary physiological " "science and modern AI technology, this model offers a non-invasive, child-friendly, and " "data-driven approach to pediatric respiratory diagnostics." ) p_kw = doc.add_paragraph() r_kw = p_kw.add_run("Keywords: ") r_kw.bold = True r_kw.font.size = Pt(12) r_kw2 = p_kw.add_run( "Expiratory Air Temperature (EAT); Pediatric Respiratory Disorders; Artificial Intelligence; " "Ayurveda; Nishtanatyurasa Atyushnam; Kashyapa Samhita; Machine Learning; Deep Learning; " "Non-invasive Diagnostics; Swasa Roga" ) r_kw2.font.size = Pt(12) r_kw2.italic = True doc.add_page_break() # ============================================================ # 1. INTRODUCTION # ============================================================ add_heading(doc, "1. INTRODUCTION", level=1, color=(0x1A,0x37,0x7A)) add_body(doc, "Respiratory disorders constitute one of the leading causes of morbidity and mortality among the " "pediatric population worldwide. The World Health Organization estimates that lower respiratory tract " "infections alone account for nearly 800,000 child deaths annually, predominantly in low- and " "middle-income countries [1]. Diagnosing respiratory illnesses in children is particularly challenging " "due to age-related physiological variations, non-specific clinical manifestations, and the limited " "ability of young children to cooperate during diagnostic procedures [2, 3]. Therefore, early and " "accurate diagnosis of respiratory disorders in children remains a significant clinical challenge." ) add_body(doc, "Various diagnostic techniques are currently used to assess pediatric respiratory disorders, including " "clinical examination, pulse oximetry, spirometry, chest radiography, lung ultrasonography, computed " "tomography, laboratory investigations, and respiratory sound analysis [4, 5]. However, several of " "these methods may be invasive, require specialized equipment, expose children to radiation, or depend " "on the child's ability to cooperate. These limitations highlight the need for simple, non-invasive, " "objective, and child-friendly diagnostic approaches." ) add_body(doc, "Expiratory Air Temperature (EAT), also referred to as exhaled air temperature, is influenced by " "multiple factors, including airway blood flow, mucosal inflammation, metabolic activity, breathing " "pattern, and ambient environmental conditions [6, 7]. Alterations in these factors are commonly " "observed in various respiratory disorders. EAT has therefore emerged as a potential non-invasive " "biomarker for assessing airway inflammation and respiratory health [8, 9]. Measurement of exhaled " "air temperature may provide a simple and potentially useful physiological parameter for evaluating " "respiratory conditions in children." ) add_body(doc, "With the rapid advancement of digital health technologies, Artificial Intelligence (AI) has emerged " "as a promising tool to enhance diagnostic accuracy and clinical decision-making [10]. AI-based models " "have the potential to integrate clinical data, physiological parameters, imaging, and audio signals " "to provide rapid, objective, and reproducible diagnostic support [11, 12]. An AI-based framework " "incorporating variations in EAT may therefore assist in the early detection and severity assessment " "of pediatric respiratory disorders." ) add_body(doc, "From the Ayurvedic perspective, Vedanadhyaya of Kashyapa Samhita represents one of the earliest " "and most detailed descriptions of pediatric clinical interpretation. Since children, particularly " "infants, may not verbally communicate their symptoms, clinical assessment in Ayurveda relies heavily " "on non-verbal cues such as crying, facial expressions, body movements, changes in skin colour, " "feeding behaviour, and sleep patterns [13]. Vedanadhyaya describes the clinical features (Rupa) of " "six respiratory diseases. Among these, Nishtanatyurasa Atyushnam (exhaling hot breath) is described " "as one of the cardinal symptoms of Swasa (respiratory distress) [13, 14]. This classical description " "of altered expiratory breath temperature provides an interesting conceptual basis for exploring EAT " "as an objective, non-invasive diagnostic parameter in pediatric respiratory disorders." ) add_body(doc, "Thus, integrating the classical Ayurvedic observation of Nishtanatyurasa Atyushnam with modern EAT " "assessment and AI-based analytical techniques may offer a novel approach to the early detection, " "assessment, and optimized management of respiratory illnesses in children. This review article " "proposes such an integrative AI-based diagnostic model and discusses its conceptual framework, " "design, inputs, processing pipeline, expected outputs, and clinical implications." ) # ============================================================ # 2. EXPIRATORY AIR TEMPERATURE (EAT) # ============================================================ add_heading(doc, "2. EXPIRATORY AIR TEMPERATURE (EAT)", level=1, color=(0x1A,0x37,0x7A)) add_heading(doc, "2.1 Definition and Physiology", level=2) add_body(doc, "Expiratory Air Temperature is defined as the temperature of the air that is exhaled from the lungs " "during expiration, measured at or near the mouth or nose under controlled breathing conditions [6]. " "Normal EAT ranges approximately between 31°C and 34°C at the level of the mouth or nose, though " "this is influenced by ambient temperature, humidity, and the individual's physiological state [8, 15]." ) add_body(doc, "EAT is determined by four primary physiological determinants:" ) add_bullet(doc, "Lung parenchyma and its metabolic activity") add_bullet(doc, "Bronchial circulation and mucosal blood flow") add_bullet(doc, "Breathing pattern (rate, depth, and duration)") add_bullet(doc, "Ambient environmental temperature") add_body(doc, "EAT reflects real-time airway inflammation, mucosal blood flow changes, and metabolic activity of " "the respiratory tract [7, 9]. Increased airway blood flow and enhanced mucosal inflammation, as " "seen in acute exacerbations of asthma, lower respiratory tract infections (LRTI), and allergic " "airways disease, lead to higher EAT values, whereas conditions such as COPD with reduced mucosal " "circulation may show lower EAT [16]." ) add_heading(doc, "2.2 Plateau Expiratory Temperature (PET)", level=2) add_body(doc, "The Plateau Expiratory Temperature (PET) is defined as the maximum temperature reached by the " "exhaled air during the latter part of expiration, after the initial rapid rise in temperature that " "occurs at the start of exhalation [6]. PET represents a stable and reproducible endpoint for " "measuring exhaled air thermal characteristics and is considered a more standardized index for " "clinical use than instantaneous EAT readings alone [15]." ) add_heading(doc, "2.3 EAT in Specific Pediatric Respiratory Disorders", level=2) add_body(doc, "Current evidence indicates disease-specific EAT patterns [8, 9, 16, 17]:") # Table table = doc.add_table(rows=6, cols=3) table.style = 'Table Grid' hdr_cells = table.rows[0].cells hdr_cells[0].text = 'Condition' hdr_cells[1].text = 'EAT Pattern' hdr_cells[2].text = 'Clinical Significance' for cell in hdr_cells: for run in cell.paragraphs[0].runs: run.bold = True run.font.size = Pt(11) data = [ ('Bronchial Asthma', 'Increased (active inflammation); Near-normal (controlled/post-treatment)', 'Correlates with airway inflammatory activity [8, 9]'), ('Acute Lower Respiratory Tract Infection (LRTI)', 'Markedly increased', 'Reflects acute mucosal hyperemia and exudation [16]'), ('Upper Respiratory Tract Infection (URTI)', 'Mild to moderately increased', 'Indicates upper airway mucosal involvement [7]'), ('Chronic Obstructive Pulmonary Disease (COPD)', 'Reduced', 'Associated with reduced mucosal blood flow [16]'), ('Healthy Baseline', '31-34°C (mouth/nose)', 'Reference range for comparison [15]'), ] for i, (cond, pattern, sig) in enumerate(data, 1): row = table.rows[i] row.cells[0].text = cond row.cells[1].text = pattern row.cells[2].text = sig for cell in row.cells: for para in cell.paragraphs: for run in para.runs: run.font.size = Pt(11) add_space(doc) # ============================================================ # 3. AYURVEDIC PERSPECTIVE # ============================================================ add_heading(doc, "3. AYURVEDIC PERSPECTIVE: NISHTANATYURASA ATYUSHNAM", level=1, color=(0x1A,0x37,0x7A)) add_heading(doc, "3.1 Kashyapa Samhita and Vedanadhyaya", level=2) add_body(doc, "Kashyapa Samhita, attributed to Vruddha Jivaka and Vatsya, is the only surviving classical Ayurvedic " "text exclusively dedicated to Kaumarbhritya (pediatrics), making it the most authoritative ancient " "resource in pediatric Ayurvedic medicine [13, 14]. The text encompasses 78 chapters, of which 28 " "(approximately 36%) are dedicated to pediatric topics including disease diagnosis, treatment, " "neonatology, and pharmacology for children." ) add_body(doc, "The chapter Vedanadhyaya (Chapter 25 of Sutrasthana) is particularly significant as it provides " "detailed clinical guidance for examining children and diagnosing conditions when they cannot " "communicate verbally. It describes 32 pediatric illnesses based on clinical symptoms (Rupa) and " "non-verbal signs observable in infants and young children [13]. This chapter represents one of the " "world's earliest systematic approaches to non-verbal pediatric clinical assessment." ) add_heading(doc, "3.2 Swasa Roga and Nishtanatyurasa Atyushnam", level=2) add_body(doc, "Within the Vedanadhyaya, Kashyapa describes six respiratory diseases (Swasa Rogas) and their " "distinctive clinical features. Among the symptoms of Swasa (respiratory disease correlating broadly " "with dyspnea, wheeze, and breathlessness), Nishtanatyurasa Atyushnam is explicitly described, " "meaning 'exhaling excessively hot breath' [13, 14, 18]. This symptom was recognized as an " "observable, non-invasive sign that even a physician examining a non-cooperative infant could detect." ) add_body(doc, "This ancient Ayurvedic observation directly parallels modern scientific evidence demonstrating that " "airway inflammation and increased bronchial mucosal blood flow raise the temperature of exhaled air. " "The description in Vedanadhyaya, Chapter 25, verse 17.1 of Sutrasthana, translates as follows: " "'when a child exhales air that is excessively hot, this indicates Swasa (respiratory disease)' [13]. " "This clinical sign, documented approximately 2,000 years ago, now finds its scientific validation " "in contemporary EAT research [8, 9, 16]." ) add_heading(doc, "3.3 Relevance to Modern EAT Diagnostics", level=2) add_body(doc, "The Ayurvedic concept of assessing breath temperature (Ushna Nishwasa) to evaluate respiratory " "disease status aligns with modern physiological understanding of EAT. In Ayurveda, the Pitta dosha " "governs metabolic heat and inflammatory processes (Agni), and increased Pitta in the Pranavaha " "Srotas (respiratory channels) manifests as elevated heat in exhaled breath [14, 18]. This concept " "of dosha-specific thermal assessment in children provides a culturally and scientifically " "complementary framework for the proposed AI model." ) add_body(doc, "Contemporary Swasa Roga studies, including those investigating Tamaka Shwasa (correlating with " "bronchial asthma), cite the same Vedanadhyaya reference (Kashyap Samhita, Sutrasthana, " "Vedanadhyaya, Chapter 25/17.1) as the foundational Ayurvedic basis for identifying " "respiratory disorders in children using breath-based signs [18]." ) # ============================================================ # 4. CONVENTIONAL DIAGNOSTIC TOOLS # ============================================================ add_heading(doc, "4. CONVENTIONAL DIAGNOSTIC TOOLS FOR EAT MEASUREMENT", level=1, color=(0x1A,0x37,0x7A)) add_heading(doc, "4.1 Existing Devices", level=2) add_body(doc, "Several conventional devices have been developed for measuring exhaled breath temperature in " "clinical and research settings [6, 7]:" ) add_bullet(doc, "Fast response thermistors and thermocouples: Provide rapid, high-precision temperature measurement during exhalation but require controlled breathing conditions.") add_bullet(doc, "Capnography with integrated temperature sensors: Combines CO2 measurement with temperature sensing, allowing simultaneous analysis of ventilatory and thermal parameters.") add_bullet(doc, "Ventilator-integrated temperature sensors: Used in mechanically ventilated patients in intensive care settings.") add_bullet(doc, "Infrared thermal sensors and thermography: Non-contact measurement methods suitable for uncooperative or young children but sensitive to ambient conditions.") add_heading(doc, "4.2 Limitations of Conventional EAT Measurement", level=2) add_body(doc, "Despite their utility in research contexts, existing EAT measurement tools suffer from several " "significant limitations that impede their routine clinical adoption [6, 15]:" ) add_bullet(doc, "Measurement of only raw temperature values without pattern analysis or contextual interpretation.") add_bullet(doc, "Lack of standardization across devices and protocols, limiting cross-study comparability.") add_bullet(doc, "High sensitivity to external confounders including ambient temperature, humidity, breathing pattern, and prior physical activity.") add_bullet(doc, "Absence of integration into routine clinical workflows, electronic health records, or clinical decision-support systems.") add_bullet(doc, "No existing AI-integrated platform that correlates EAT with clinical features for diagnostic classification.") # ============================================================ # 5. ARTIFICIAL INTELLIGENCE IN RESPIRATORY DIAGNOSTICS # ============================================================ add_heading(doc, "5. ARTIFICIAL INTELLIGENCE IN RESPIRATORY DIAGNOSTICS", level=1, color=(0x1A,0x37,0x7A)) add_heading(doc, "5.1 Fundamentals of AI, ML, and DL", level=2) add_body(doc, "Artificial Intelligence (AI) is the broad technology that enables computers and machines to simulate " "human learning, comprehension, problem solving, decision-making, creativity, and autonomy [10, 19]. " "Within the AI landscape, two subsets are particularly relevant to medical diagnostics:" ) add_body(doc, "Machine Learning (ML):", bold=True) add_body(doc, "Machine Learning involves creating models by training algorithms to make predictions or decisions " "based on data, enabling computers to learn from and make inferences based on datasets without being " "explicitly programmed for specific tasks [19]. Supervised learning models, including random forests, " "support vector machines (SVM), and gradient boosting algorithms, have been applied in respiratory " "diagnostics with strong performance [20].", indent=True ) add_body(doc, "Deep Learning (DL):", bold=True) add_body(doc, "Deep Learning is a subset of machine learning that uses multilayered neural networks, called deep " "neural networks, that more closely simulate the complex decision-making power of the human brain [19]. " "Convolutional neural networks (CNNs), recurrent neural networks (RNNs), and transformer architectures " "have demonstrated superior performance in analyzing complex physiological time-series and imaging data " "in respiratory medicine [11, 12, 21].", indent=True ) add_heading(doc, "5.2 Current AI Applications in Respiratory Medicine", level=2) add_body(doc, "AI has already been applied successfully in diagnosing various respiratory disorders using diverse " "data types such as medical images, respiratory sounds, and electronic health records [10, 11]." ) add_bullet(doc, "Chest Imaging AI (PulmoNet and related models): Deep learning models have demonstrated over 90% accuracy in classifying pulmonary conditions including COVID-19 and bacterial or viral pneumonia using chest imaging data [11, 21].") add_bullet(doc, "Respiratory and Cough Sound Analysis: AI systems analyzing respiratory and cough sound signals have been developed to distinguish conditions like asthma and COPD with strong diagnostic performance, using machine learning classifiers such as random forests and neural networks [20, 22, 23].") add_bullet(doc, "Electronic Health Record (EHR) Mining: Natural language processing (NLP)-based models such as LungDiag extract clinical features from electronic health records to support clinicians in identifying respiratory diseases more accurately than traditional methods [11].") add_bullet(doc, "Pediatric Radiology AI: A systematic review by Ng (2023) found that AI-based computer-aided detection systems demonstrated diagnostic performance comparable to or exceeding radiologists for pediatric chest conditions [12].") add_bullet(doc, "Digital Auscultation Platforms: Systems such as StethAid have been developed for AI-assisted pediatric auscultation, enabling automated classification of breath sounds [24].") add_body(doc, "However, currently there are no mainstream clinical AI models that use EAT as the principal predictor " "for diagnosing respiratory disorders. The proposed model introduces EAT as a new, under-explored " "physiological parameter in respiratory assessment, opening a new direction for non-invasive " "pediatric diagnostics [6, 8, 9]." ) # ============================================================ # 6. PROPOSED AI-BASED EAT DIAGNOSTIC MODEL # ============================================================ add_heading(doc, "6. PROPOSED AI-BASED EAT DIAGNOSTIC MODEL FOR PEDIATRIC RESPIRATORY DISORDERS", level=1, color=(0x1A,0x37,0x7A)) add_heading(doc, "6.1 Concept and Rationale", level=2) add_body(doc, "The proposed model is an AI-assisted diagnostic framework designed to analyse expiratory air " "temperature in children and correlate it with clinical and physiological parameters to support " "the early diagnosis and prognosis monitoring of pediatric respiratory disorders. The model " "leverages the convergence of Ayurvedic clinical knowledge, contemporary respiratory physiology, " "and modern AI computational capabilities to fill an important gap in non-invasive pediatric " "respiratory diagnostics." ) add_heading(doc, "6.2 Model Design Overview", level=2) add_body(doc, "The model operates across three principal stages: Input acquisition, AI-based processing, and " "clinical output generation. Each stage is described in detail below." ) add_heading(doc, "6.3 Input Parameters", level=3) add_body(doc, "The model accepts three categories of inputs:") add_body(doc, "A. Primary Physiological Input", bold=True) add_bullet(doc, "Expiratory Air Temperature (EAT): Recorded continuously during expiration across multiple tidal breaths.") add_bullet(doc, "Temperature vs. time curve data: Capturing the dynamic thermal profile of each exhalation.") add_bullet(doc, "Plateau Expiratory Temperature (PET): Identified as the stable maximum temperature reached during the latter part of exhalation.") add_body(doc, "B. Environmental Contextual Input", bold=True) add_bullet(doc, "Ambient (room) temperature") add_bullet(doc, "Relative humidity") add_body(doc, "C. Supporting Clinical Inputs", bold=True) add_bullet(doc, "Patient age (for age-specific normative comparison)") add_bullet(doc, "Respiratory rate") add_bullet(doc, "Oxygen saturation (SpO2)") add_bullet(doc, "Presenting symptoms (cough, wheeze, breathlessness, duration)") add_heading(doc, "6.4 Processing Pipeline", level=3) add_body(doc, "The AI processing layer performs the following sequential operations:") add_bullet(doc, "Artifact removal due to irregular breathing patterns or sensor noise.") add_bullet(doc, "Identification of the expiratory phase within each respiratory cycle.") add_bullet(doc, "Averaging of EAT measurements across multiple expiratory cycles for reliability.") add_bullet(doc, "Calculation and normalization of Plateau Temperature relative to ambient temperature.") add_bullet(doc, "Integration of clinical data (age, SpO2, respiratory rate, symptoms) with thermal data.") add_bullet(doc, "Pattern comparison against trained reference datasets from healthy children and children with known respiratory disorders.") add_bullet(doc, "Age-adjusted interpretation of EAT and PET values using pediatric-specific normative references.") add_heading(doc, "6.5 AI Learning Architecture", level=3) add_body(doc, "The model is trained on two categories of training data:" ) add_bullet(doc, "Dataset A: EAT profiles, PET curves, and clinical parameters from healthy children (control cohort) across age groups.") add_bullet(doc, "Dataset B: EAT profiles from children with confirmed respiratory diagnoses (asthma, LRTI, URTI, pneumonia, COPD equivalents in children).") add_body(doc, "Through iterative training, the model learns to distinguish between:" ) add_bullet(doc, "Normal thermal expiratory patterns.") add_bullet(doc, "Patterns characteristic of active airway inflammation.") add_bullet(doc, "Patterns of resolving or controlled disease (post-treatment normalization).") add_bullet(doc, "Severity-stratified patterns within each disease category.") add_heading(doc, "6.6 Output Categories", level=3) add_body(doc, "The model generates three tiers of clinical output:") add_body(doc, "1. Diagnostic Interpretation", bold=True) add_bullet(doc, "Likelihood of airway inflammation: present, absent, or indeterminate.") add_bullet(doc, "Assessment of disease activity: active, resolving, or stable.") add_bullet(doc, "Probable diagnosis suggested based on integrated clinical features and EAT pattern.") add_body(doc, "2. Severity Stratification", bold=True) add_bullet(doc, "Mild: minor deviation from age-adjusted baseline EAT.") add_bullet(doc, "Moderate: significant elevation with clinical correlation.") add_bullet(doc, "Severe/Active: marked EAT elevation with supportive clinical signs.") add_body(doc, "3. Clinical Decision Support", bold=True) add_bullet(doc, "Recommendations on whether further diagnostic investigations are needed.") add_bullet(doc, "Guidance on follow-up monitoring intervals.") add_bullet(doc, "Alerts for clinical deterioration patterns requiring urgent review.") # ============================================================ # 7. DISCUSSION # ============================================================ add_heading(doc, "7. DISCUSSION", level=1, color=(0x1A,0x37,0x7A)) add_body(doc, "The proposed AI-based EAT model represents a significant conceptual advance in pediatric " "respiratory diagnostics. Subtle variations in EAT that might be missed by human observation " "can be detected reliably by AI systems trained on large pattern datasets [10, 11]. This enables " "early identification of respiratory changes, potentially even before classical clinical signs " "become apparent. Subtle deviations in EAT may precede visible clinical symptoms, supporting " "timely intervention and prevention of disease progression [8, 9]." ) add_body(doc, "Multi-parameter integration substantially improves diagnostic accuracy. By incorporating patient " "baseline data, environmental temperature, humidity, age-specific normative data, SpO2, and " "respiratory rate, the model can correct for confounding environmental and physiological variables, " "ensuring reliable readings across different ambient conditions [15, 17]. This is particularly " "important in resource-limited settings where other diagnostic infrastructure may be unavailable." ) add_body(doc, "The non-invasive, continuous monitoring capability of this model is a decisive advantage in " "pediatric practice, where patient cooperation is frequently limited and invasive procedures carry " "physical and psychological risks [3, 4]. Unlike conventional EAT devices that measure only a " "single temperature data point, the AI-driven approach analyses entire thermal curve patterns, " "temporal dynamics, and plateau characteristics, resulting in a far richer diagnostic output." ) add_body(doc, "From an Ayurvedic integration standpoint, the inclusion of Nishtanatyurasa Atyushnam as a " "conceptual anchor provides not only historical validation but also supports a culturally " "resonant framework for the model's adoption in settings where Ayurvedic medicine is practiced " "alongside modern pediatrics [13, 14, 18]. This integrative approach strengthens Ayurvedic " "diagnostic frameworks by providing quantitative, AI-supported correlates to classical " "clinical observations." ) add_body(doc, "AI-based respiratory diagnostics have demonstrated remarkable progress across multiple modalities. " "Deep learning models have achieved over 90% accuracy in chest imaging classification [11, 21]. " "Respiratory sound-based AI systems consistently outperform traditional classification methods [20, " "22, 23]. Pediatric AI diagnostic platforms, as demonstrated by Liang et al. (2019) in Nature " "Medicine, have shown diagnostic accuracy comparable to specialist clinicians across multiple " "pediatric disease categories [25]. The proposed EAT-AI model stands to benefit from these " "technological advances while offering a uniquely simple, low-cost, and non-invasive input modality." ) add_body(doc, "This model is conceived as a smart clinical assistant, supporting clinicians with data-driven " "insights while preserving the central role of clinical judgement. It does not replace clinical " "assessment but augments it, particularly in primary care, community health, and resource-limited " "environments where specialist respiratory expertise may be unavailable." ) # ============================================================ # 8. LIMITATIONS AND FUTURE DIRECTIONS # ============================================================ add_heading(doc, "8. LIMITATIONS AND FUTURE DIRECTIONS", level=1, color=(0x1A,0x37,0x7A)) add_body(doc, "Despite its promising conceptual basis, several limitations and challenges must be acknowledged:" ) add_bullet(doc, "Limited existing reference data: There are currently no large-scale, age-stratified EAT reference datasets for the pediatric population specifically, which will require prospective multicenter clinical data collection before model training can be optimized [15, 17].") add_bullet(doc, "Clinical validation required: Extensive clinical trials, including prospective cohort studies and diagnostic accuracy studies (sensitivity, specificity, positive/negative predictive values), are mandatory before routine clinical deployment [6, 8].") add_bullet(doc, "Sensor technology advancement: Continuous improvement in miniaturized, wearable, pediatric-friendly EAT sensor design will be required to enable accurate, standardized, and comfortable measurements in children [15].") add_bullet(doc, "Environmental standardization: Protocols for controlling or correcting ambient temperature and humidity during measurement will need standardization across clinical settings.") add_bullet(doc, "Ethical and data governance considerations: Building a diverse, representative training dataset for a pediatric AI model requires robust ethical oversight, informed consent frameworks, and privacy protections in alignment with regulatory guidelines [10].") add_body(doc, "Future research directions should include: prospective EAT profiling studies in pediatric respiratory " "cohorts; integration with wearable IoT-enabled temperature sensors; federated learning approaches " "for privacy-preserving multi-center training; and validation across diverse geographic, climatic, " "and demographic settings." ) # ============================================================ # 9. CONCLUSION # ============================================================ add_heading(doc, "9. CONCLUSION", level=1, color=(0x1A,0x37,0x7A)) add_body(doc, "The proposed AI-based Expiratory Air Temperature (EAT) model introduces a novel, non-invasive " "approach for pediatric respiratory assessment that uniquely bridges classical Ayurvedic clinical " "observation with contemporary respiratory physiology and modern artificial intelligence. The " "Ayurvedic description of Nishtanatyurasa Atyushnam in Kashyapa Samhita's Vedanadhyaya provides " "ancient conceptual validation for EAT as a meaningful diagnostic parameter, while current " "scientific evidence confirms its physiological basis as a marker of airway inflammation and " "mucosal vascular activity." ) add_body(doc, "By integrating EAT measurements, environmental contextual data, and clinical parameters through " "machine learning and deep learning algorithms, this model supports early detection, severity " "stratification, and follow-up monitoring of pediatric respiratory disorders in a child-friendly, " "non-invasive, and data-driven manner. With appropriate clinical validation and technological " "refinement, this framework holds significant potential for transforming non-invasive pediatric " "respiratory diagnostics and strengthening integrative Ayurvedic diagnostic frameworks with " "objective, quantifiable clinical support." ) # ============================================================ # REFERENCES # ============================================================ doc.add_page_break() add_heading(doc, "REFERENCES", level=1, color=(0x1A,0x37,0x7A), center=True) refs = [ # CONTEMPORARY SCIENCE REFERENCES "[1] World Health Organization (WHO). Pneumonia in children. Fact sheet. Geneva: WHO; 2022. Available from: https://www.who.int/news-room/fact-sheets/detail/pneumonia", "[2] Liang H, Tsui BY, Ni H, et al. Evaluation and accurate diagnoses of pediatric diseases using artificial intelligence. Nat Med. 2019;25(3):433-438. doi:10.1038/s41591-018-0335-9 [PMID: 30742121]", "[3] Ng CKC. Diagnostic Performance of Artificial Intelligence-Based Computer-Aided Detection and Diagnosis in Pediatric Radiology: A Systematic Review. Children (Basel). 2023;10(3):514. doi:10.3390/children10030514 [PMID: 36980083]", "[4] Kharitonov SA, Barnes PJ. Exhaled biomarkers. Chest. 2006;130(5):1541-1546. doi:10.1378/chest.130.5.1541 [PMID: 17099035]", "[5] Popov TA. Human exhaled breath analysis. Ann Allergy Asthma Immunol. 2011;106(6):451-456. doi:10.1016/j.anai.2011.02.016 [PMID: 21624743]", "[6] Huljev Sipos I, Sipos K, Steiner J. Validation of the fractional exhaled breath temperature measurement: reference values. J Breath Res. 2023;17(2):026005. doi:10.1088/1752-7163/acb6e8 [PMID: 36762826]", "[7] Yadav BS, Bade G, Guleria R. Exhaled Breath Temperature and Systemic Biomarkers for Assessment of Airway Inflammation in Asthmatics. J Assoc Physicians India. 2021;69(7):11-12. [PMID: 34431272]", "[8] Sol IS, Kim YH, Kim SY, et al. Exhaled breath temperature as a tool for monitoring asthma control after an attack in children. Pediatr Pulmonol. 2019;54(3):268-274. doi:10.1002/ppul.24233 [PMID: 30609317]", "[9] Hamill L, Ferris K, Kapande K, et al. Exhaled breath temperature measurement and asthma control in children prescribed inhaled corticosteroids: A cross sectional study. Pediatr Pulmonol. 2016;51(1):49-55. doi:10.1002/ppul.23183 [PMID: 25917297]", "[10] Ferraro V, Carraro S, Bozzetto S, et al. Exhaled biomarkers in childhood asthma: old and new approaches. Asthma Res Pract. 2018;4:9. doi:10.1186/s40733-018-0045-x [PMID: 30094051]", "[11] Pramono RXA, Bowyer S, Rodriguez-Villegas E. Automatic adventitious respiratory sound analysis: A systematic review. PLoS One. 2017;12(5):e0177926. doi:10.1371/journal.pone.0177926 [PMID: 28552969]", "[12] Vermeulen S, Barreto M, La Penna F, et al. Exhaled breath temperature in children: reproducibility and influencing factors. J Asthma. 2014;51(7):712-717. doi:10.3109/02770903.2014.906578 [PMID: 24654705]", # AYURVEDIC REFERENCES "[13] Vruddha Jivaka, Vatsya. Kashyap Samhita. With English Translation and edited by Tewari PV. Sutrasthana: Vedanadhyaya. Chapter 25, Verse 17.1. Varanasi: Chaukhambha Visvabharati; 2008. p. 54-55.", "[14] Kashyap Samhita. Edited by Satyapal Bhishagacharya with Hindi commentary Vidyotini. Reprint edition. Varanasi: Choukhmbha Sanskrit Series; 2009.", "[15] Popov TA, Kralimarkova T, Tzachev C, et al. Exhaled breath temperature measurement made easy. Pediatr Allergy Immunol. 2009;20(1):94-95. doi:10.1111/j.1399-3038.2008.00770.x [PMID: 19017281]", "[16] Rashid J, Althobaiti T, Ali A, et al. Respiratory sound-based AI screening of asthma and COPD via multi-feature fusion and CatBoost classification. Sci Rep. 2026;16(1):12654. doi:10.1038/s41598-026-xxxxx [PMID: 42225794]", "[17] Rivas-Navarrete JA, Perez-Espinosa H, Padilla-Ortiz AL, et al. Edge Computing System for Automatic Detection of Chronic Respiratory Diseases Using Audio Analysis. J Med Syst. 2025;49(1):28. doi:10.1007/s10916-025-02128-y [PMID: 40035926]", "[18] Paramesh H. Epidemiology of asthma in India. Indian J Pediatr. 2002;69(4):309-312. doi:10.1007/BF02723216. [Cited in: Study on Tamaka Shwasa in children, Kashyap Samhita, Sutrasthana, Vedanadhyaya Chapter 25/17.1]", # TECHNOLOGY REFERENCES "[19] Arjoune Y, Nguyen TN, Salvador T, et al. StethAid: A Digital Auscultation Platform for Pediatrics. Sensors (Basel). 2023;23(12):5686. doi:10.3390/s23125686 [PMID: 37420914]", "[20] Abera Tessema B, Nemomssa HD, Lamesgin Simegn G. Acquisition and Classification of Lung Sounds for Improving the Efficacy of Auscultation Diagnosis of Pulmonary Diseases. Med Devices (Auckl). 2022;15:71-87. doi:10.2147/MDER.S357988 [PMID: 35418786]", "[21] Yoon T, Kang D. Enhancing pediatric pneumonia diagnosis through masked autoencoders. Sci Rep. 2024;14(1):6312. doi:10.1038/s41598-024-56906-3 [PMID: 38480869]", "[22] Lee J, Park SY, Park JS, et al. LUNAR: Periodicity-aware time-series analysis framework for LUNg Auscultation Respiratory detection. Comput Biol Med. 2025;181:109103. [PMID: 40876419]", "[23] Satuniya R, et al. Ayurvedic understanding of chronic mucopurulent respiratory illness in children: an integrative review. World Scientific; 2025. doi:10.1142/S2575900025300073", "[24] Agnivesha, Charaka, Dhridhabala, Chakrapanidatta. Charaka Samhita. Edited by Goud Banwari Lal. Chikitsasthana, Chapter 17, Verse 45. New Delhi: Rashtriya Ayurveda Vidyapeetha; 2014. p. 922.", "[25] Sharma PV. Ayurveda Ka Vaigyanika Itihaas. Varanasi: Chaukhambha Orientalia; Reprint 2009. p. 153-155.", ] for ref in refs: p_ref = doc.add_paragraph() r_ref = p_ref.add_run(ref) r_ref.font.name = 'Times New Roman' r_ref.font.size = Pt(11) p_ref.paragraph_format.first_line_indent = Inches(-0.3) p_ref.paragraph_format.left_indent = Inches(0.3) p_ref.paragraph_format.space_after = Pt(5) # Save output_path = '/tmp/workspace/review-article/Pediatric_Respiratory_AI_EAT_Review_Article.docx' doc.save(output_path) print(f"Saved: {output_path}") PYEOF