Daily Respiratory Research Analysis
Analyzed 59 papers and selected 3 impactful papers.
Summary
Today's top respiratory research spans AI-augmented pathology, precision sleep medicine phenotyping, and population-tailored risk screening. An AI system across 20 hospitals dramatically accelerated and improved nasal polyp subtype diagnosis and recurrence prediction; a MESA cohort analysis showed that OSA patients with high hypoxic burden or heart-rate surges have higher incident AF risk; and a large Japanese study introduced the HANDSOME score to screen Asians for sleep-disordered breathing.
Research Themes
- AI-assisted pathology and prognostics in airway disease
- Precision phenotyping in sleep-disordered breathing to predict cardiovascular outcomes
- Population-tailored screening tools for respiratory conditions
Selected Articles
1. Artificial Intelligence-Based Pathological Subtype Diagnosis of Nasal Polyps: A Multidimensional and Micro-Visualization Study.
Across 20 hospitals and 2,457 slides, an AI system combining cell detection, region segmentation, and 3D reconstruction achieved high accuracy for nasal polyp subtype diagnosis, dramatically reducing reading time and improving junior pathologist performance. A WSI-based prognostic model (AUC 86.6%) outperformed an MI-based model, indicating added clinical utility.
Impact: This study demonstrates a rigorously validated, multi-institutional AI platform that improves diagnostic accuracy, speed, and prognostication for a common airway disease, potentially reshaping pathology workflows.
Clinical Implications: Pathology departments can adopt AI-augmented workflows to standardize nasal polyp subtype diagnosis, reduce turnaround time, and incorporate prognostic modeling to inform recurrence risk and individualized therapy.
Key Findings
- Weighted F1-score 0.809 for cell detection and IoU 0.827 for region segmentation on internal data; external F1 0.792 and IoU 0.815
- Diagnosis accuracy reached 90% (MI) and 91% (WSI), with time reduction from 193 to 8 s (MI) and 10,450 to 250 s (WSI)
- Junior pathologists improved accuracy from 50% to 89% with AI support
- WSI-based prognostic model achieved AUC 86.64% and outperformed MI-based model (AUC 79.81%, p=0.039)
Methodological Strengths
- Large multi-center dataset (2,457 slides from 20 hospitals) with external validation
- Reader study involving 12 pathologists and integrated 3D spatial quantification
Limitations
- Clinical impact assessed in retrospective reader studies; prospective real-world implementation not reported
- Recurrence prediction trained on a single 131-patient cohort; generalizability to broader populations requires validation
Future Directions: Prospective, multi-center clinical trials to evaluate workflow integration, patient outcomes, and cost-effectiveness; expansion to other airway inflammatory diseases and federated learning for privacy-preserving model updates.
BACKGROUND: Nasal polyps (NP) are common upper respiratory conditions with diverse inflammatory subtypes influencing clinical features and prognosis. Manual counting of inflammatory cells in microscopic images (MI) is laborious and subjective, limiting diagnostic precision and treatment decisions. METHODS: A total of 2457 slides from 20 hospitals were used to develop an AI-based NP subtype diagnosis system (NPSS). NPSS-MI was built using 1047 slides (15,705 MIs) annotated by pathologists. NPSS-WSI was trained on 1410 slides (21,150 images) combining PA-P2PNet for cell detection and U-KAN for region segmentation. Three-dimensional reconstruction (3DNP) using registration and point cloud analysis enabled spatial quantification of inflammatory cells. Twelve pathologists evaluated NPSS accuracy and efficiency on 200 slides, and recurrence prediction models were developed using logistic regression in a 131-patient cohort. RESULTS: NPSS achieved performance with a weighted average F1-score of 0.809 for cell detection and an intersection over union (IoU) of 0.827 for region segmentation with an internal dataset. External dataset performance showed an F1-score of 0.792 and an IoU of 0.815. Forty randomly accumulated MIs were approximated WSI results. NPSS-MI and NPSS-WSI reached accuracies of 90% and 91%, reducing diagnostic time from 193 to 8 s and from 10,450 to 250 s, respectively. Junior pathologists using NPSS improved accuracy from 50% to 89%. Inflammatory cells showed distinct spatial patterns in 3DNP. NPSS-WSI prognostic model outperformed the MI-based model (AUC 86.64% vs. 79.81%, p = 0.039). CONCLUSIONS: NPSS integrates MI, WSI, and 3DNP to enable accurate and efficient NP subtype diagnosis and prognosis prediction, greatly enhancing diagnostic precision and clinical utility.
2. Association of High Cardiovascular Disease Risk Obstructive Sleep Apnea with Incident Atrial Fibrillation: the Multi-Ethnic Study of Atherosclerosis.
In a multi-ethnic cohort, OSA patients with high hypoxic burden or high heart rate response to respiratory events had a significantly higher risk of incident atrial fibrillation compared with non-OSA, whereas OSA without these features did not. Findings were consistent across sexes and support precision phenotyping beyond AHI.
Impact: Links mechanistically relevant OSA phenotypes (hypoxemia and autonomic response) to AF risk, offering a clinically practical way to refine cardiovascular risk stratification in sleep clinics.
Clinical Implications: Incorporating hypoxic burden and heart-rate response into OSA assessment may better identify patients who benefit from aggressive cardiovascular risk management and targeted OSA therapies.
Key Findings
- Among 1,679 participants, AF incidence was 14.1% (High-CVD-Risk OSA), 10.1% (Low-CVD-Risk OSA), and 8.4% (Non-OSA) over 6.7 years
- High-CVD-Risk OSA had increased AF risk vs. Non-OSA (HR 1.68; 95% CI 1.17–2.41); Low-CVD-Risk OSA did not (HR 1.20; 95% CI 0.76–1.92)
- Risk associations were similar in women and men (HR 1.71 vs 1.64, respectively)
Methodological Strengths
- Prospective cohort analysis within a well-characterized multi-ethnic study with median 6.7-year follow-up
- Use of physiologically grounded metrics (hypoxic burden and ΔHR) and multivariable Cox models
Limitations
- Observational design limits causal inference and residual confounding may persist
- Operational definitions (tertile cutoffs) for hypoxic burden and ΔHR require external replication and clinical thresholds
Future Directions: Prospective interventional studies testing whether targeting high hypoxic burden/ΔHR reduces AF incidence; integration into clinical decision tools and validation across diverse populations.
BACKGROUND: Obstructive sleep apnea (OSA) is associated with atrial fibrillation (AF), but this association varies by population and definitions of OSA and AF. RESEARCH QUESTION: Whether 'high cardiovascular disease risk' (High-CVD-Risk) OSA, defined by elevated hypoxic burden (HB) or heart rate response to respiratory events (ΔHR), is associated with incident AF. STUDY DESIGN AND METHODS: Using data from the Multi-Ethnic Study of Atherosclerosis, HB was quantified as the cumulative area under the desaturation curve following respiratory events, and ΔHR as the increase in heart rate upon event termination. Within OSA (apnea-hypopnea index (AHI) ≥15 events/h), High-CVD-Risk OSA was defined as having either a high ΔHR or high HB (both in highest tertile), while individuals with OSA who did not meet these criteria were considered as Low-CVD-Risk. Cox proportional hazards models were used to estimate the adjusted hazard ratios (HR) of incident AF for High- and Low-CVD-Risk OSA (vs Non-OSA, defined as an AHI < 15 events/h) after adjusting for confounders. RESULTS: A total of 1,679 participants (45.4% male) were included, with a median [interquartile range] age of 66.0 [60.0-74.0] years, HB of 35.3 [18.0-69.2] %min/h, and ΔHR of 7.7 [5.9-9.9] bpm. During a median follow-up of 6.7 years, AF was identified in 14.1% of the High-CVD-Risk OSA (n = 687), 10.1% of the Low-CVD-Risk OSA (n = 278), and 8.4% of the Non-OSA (n = 714). Compared to the Non-OSA, an increased hazard ratio for AF was found in High-CVD-Risk OSA (HR, 1.68; 95% CI,1.17-2.41) but not in Low-CVD-Risk OSA (HR, 1.20; 95% CI, 0.76-1.92). The adjusted hazard ratio was similar in women and men ([HR, 1.71; 95% CI, 1.01-2.91] vs [HR, 1.64; 95% CI, 0.99-2.72]). INTERPRETATION: OSA-related hypoxemia or heart rate surge may be useful for predicting which patients with OSA are at increased risk for incident AF.
3. Novel predictive score for sleep disordered breathing in Asians: the Nagahama study.
Using over 7,700 community participants, the Nagahama study derived and validated the HANDSOME score for moderate-to-severe SDB in Asians, leveraging objective actigraphy-based ODI metrics. Independent predictors included hypertension, age ≥60, nocturia, diabetes, snoring, and obesity.
Impact: Provides a pragmatic, objective, and population-tailored SDB screening tool for Asians, addressing performance gaps of existing questionnaires.
Clinical Implications: Primary care and sleep clinics in Asia can use HANDSOME to prioritize diagnostic testing (e.g., HSAT/PSG) and guide early interventions for high-risk individuals.
Key Findings
- Discovery (n=6,375) and validation (n=1,338) cohorts from a community-based Japanese study
- Moderate-to-severe SDB defined by actigraphy-corrected 3% ODI ≥15/h across ≥2 nights
- Independent predictors: hypertension, age ≥60 years, nocturia frequency, diabetes, reported snoring, and obesity (BMI thresholds)
Methodological Strengths
- Large discovery and independent validation cohorts with objective actigraphy-derived ODI metrics
- Weighted predictor scoring tailored to Asian populations
Limitations
- Cross-sectional design limits causal inferences and temporal stability of the score
- Actigraphy-based ODI may differ from polysomnography-derived metrics; external validation beyond Japan is needed
Future Directions: Prospective studies to assess impact on diagnostic yield and outcomes; calibration for different Asian subpopulations; integration with wearable-derived physiology for continuous risk monitoring.
BACKGROUND: Early detection of sleep disordered breathing (SDB) is limited by the insufficient accuracy of current screening tools, especially for Asians. This study aimed to clarify factors associated with SDB and develop an objective risk score for Asian populations. METHODS: We conducted a cross-sectional analysis of community participants from the Nagahama study in Japan, including 6375 people in the discovery cohort and 1338 people in the validation cohort. SDB was assessed by the 3% oxygen desaturation index corrected for sleep duration obtained by wrist actigraphy (Acti-ODI3%) over ≥2 nights, and moderate-to-severe SDB was defined as Acti-ODI3% ≥ 15/h. A risk scoring system was developed using weighted predictors. RESULTS: Independent predictors of moderate-to-severe SDB included Hypertension (1 point), older Age (≥60 years; 2 point), Nocturia frequency (1 to <2/≥2 times/day; 1/2 points), Diabetes (1 point), reported Snoring (2 points), Obesity (body mass index 22.5 to <25/≥25 kg/m CONCLUSIONS: The HANDSOME score is a simple and practical tool tailored for Asian populations that effectively identifies individuals at high risk of moderate-to-severe SDB.