Masahiro Yanagawa

Purpose: To investigate the association between AI-derived probability of interstitial lung abnormalities (ILA), including interstitial lung disease (ILD), and prognosis in an esophageal cancer cohort re-evaluated according to the 2025 American Thoracic Society (ATS) Clinical Statement. Materials and methods: This retrospective study included patients with esophageal cancer who underwent pretreatment chest CT between January 2011 and December 2015. Chest CT images were visually assessed and classified as non-ILA, indeterminate for ILA, ILA, or ILD according to the 2025 ATS Clinical Statement. Twelve previously developed artificial intelligence (AI) models generated patient-level ILA probability scores ranging from 0 to 1. Associations between AI-derived ILA/ILD probability and five-year overall survival (OS) were evaluated using multivariable Cox proportional hazards models adjusted for age, sex, body mass index, smoking history, and histology, stratified by clinical stage. Associations between visual ILA/ILD classification and OS were also assessed. Results: A total of 452 patients (median age, 68 years [interquartile range: 61–73 years]; 61 women) were included. Among them, 251 (56%) were classified as non-ILA, 118 (26%) as indeterminate for ILA, 75 (17%) as ILA, and 8 (2%) as ILD. All 12 AI models showed significant associations with shorter OS (all p < 0.05). The best-fit model yielded a hazard ratio (HR) for mortality of 7.18 (95% confidence interval [CI]: 2.70–19.1; p < 0.001). Visual ILA and ILD were associated with mortality (ILA: HR = 1.60, 95% CI: 1.02–2.51, p = 0.039; ILD: HR = 3.01, 95% CI: 1.07–8.44, p = 0.037). AI-derived probability scores differed significantly among visual categories (all p < 0.01). Conclusion: AI-derived ILA/ILD probability was associated with five-year OS in patients with esophageal cancer and may provide a reproducible quantitative marker for opportunistic risk stratification using chest CT.

Intermuscular adipose tissue (IMAT) refers to adipose tissue located between muscle groups or beneath the muscle fascia, and its accumulation is closely linked to impaired muscle quality, reduced contractile function, and adverse metabolic profiles, establishing it as a core component of myosteatosis. Unlike subcutaneous and visceral adipose tissue, which reside outside the muscle architecture, IMAT expands within the muscle environment, where local metabolic and mechanical interactions may affect muscle integrity and performance. Elevated IMAT has been associated with diminished strength, impaired mobility, frailty, insulin resistance, and adverse cardiometabolic profiles, underscoring its clinical relevance beyond a passive fat depot. However, IMAT evidence remains dispersed across imaging techniques, anatomical regions, and disease‐specific literatures, limiting a unified understanding of its clinical utility. This review synthesizes current evidence on IMAT biology, imaging‐based quantification, disease‐specific distribution and clinical implications, with emphasis on its value as a risk‐stratification biomarker. Recent advances in medical imaging, including computed tomography (CT), magnetic resonance imaging (MRI), ultrasound and positron emission tomography (PET), have enabled noninvasive and reproducible IMAT quantification. These modalities provide complementary measures of muscle fat infiltration, including CT‐derived area, volume and attenuation; MRI‐based segmentation and proton density fat fraction; ultrasound echo intensity; and PET‐derived metabolic activity. Deep learning–based segmentation methods have improved large‐scale IMAT quantification and reproducibility across anatomical regions. Mechanistically, IMAT is associated with chronic low‐grade inflammation and impaired glucose homeostasis, promoting progressive fat infiltration and muscle degeneration. Human studies indicate that IMAT exhibits an inflammatory secretory profile linked to insulin resistance and metabolic dysfunction. Quantitative imaging shows anatomically heterogeneous IMAT distribution, with frequent involvement of the lower extremities, paraspinal muscles and trunk. Across disease settings, IMAT appears to provide complementary information beyond muscle mass alone by capturing muscle quality and systemic vulnerability. Evidence is strongest in oncology, where elevated IMAT or IMAT‐based ratios have repeatedly been associated with poorer survival, recurrence, treatment toxicity, and reduced treatment tolerance. In metabolic and musculoskeletal disorders, IMAT is consistently related to insulin resistance, reduced strength, impaired mobility, and frailty, whereas cardiovascular evidence suggests prognostic relevance but remains less standardized and more heterogeneous. Clinically, IMAT imaging may help identify patients whose preserved muscle size or body mass index masks poor muscle quality, thereby refining prognostic assessment, perioperative risk evaluation, and selection for nutritional, exercise, metabolic, or treatment strategies. Future efforts should focus on standardized acquisition, automated segmentation, harmonized thresholds, and prospective validation to support integration of IMAT metrics into clinical decision‐making.

Background: We investigated whether radiomic scores (RadScores), computed from high‐dimensional quantitative features extracted from cardiac magnetic resonance parametric mapping images, can predict left ventricular reverse remodeling (LVRR) in patients with nonischemic dilated cardiomyopathy. Methods: This retrospective analysis included 74 patients with nonischemic dilated cardiomyopathy who received optimized medical therapy and underwent cardiac magnetic resonance. LVRR was defined as a left ventricular ejection fraction ≥40% with ≥10% increase at least 180 days after cardiac magnetic resonance. RadScores were calculated for each case as linear combinations of selected radiomic features, with coefficients derived from least absolute shrinkage and selection operator–penalized logistic regression on native longitudinal relaxation time (T1) and extracellular volume (ECV) fraction maps. Predictive performance for LVRR was evaluated using logistic regression and receiver operating characteristic analyses. Results: Patients with LVRR (n=31, 42%) had higher T1‐RadScore (0.72±1.52 versus −1.54±2.66; P P =0.005), lower mean‐ECV (31.5%±3.8% versus 34.9%±6.0%; P =0.004), and lower extent of late gadolinium enhancement (0.19±0.58 versus 2.99±6.5; P =0.047) than those without LVRR. The area under the receiver operating characteristic curve for predicting LVRR was 0.834 (95% CI, 0.716–0.909) for the T1‐RadScore and 0.718 (95% CI, 0.573–0.827) for the ECV‐RadScore. Adding T1‐ and ECV‐RadScore combination to the extent of late gadolinium enhancement model significantly improved the area under the receiver operating characteristic curve from 0.618 to 0.903 ( P <0.0001). Conclusions: Radiomic scores extracted from native T1 and ECV maps may enhance LVRR prediction and provide incremental value beyond traditional late gadolinium enhancement assessment in patients with dilated cardiomyopathy.

Purpose: To evaluate how two fixed computer-aided detection (CAD) operating modes—sensitivity-optimized (SE-CAD) and specificity-optimized (SP-CAD)—affect radiologists’ diagnostic performance in chest radiography (CXR). Previous studies have focused on the performance of standalone CAD at different threshold settings. However, in clinical practice, CAD is commonly used as a decision-support tool for radiologists. This study aimed to evaluate the impact of different CAD threshold settings on radiologists’ diagnostic performance. Materials and methods: This retrospective single-center diagnostic accuracy study evaluated posteroanterior or anteroposterior CXRs acquired between 2013 and 2023, interpreted with and without AI assistance, using CT as the reference standard. Six radiologists (three residents / three board-certified) read all cases four times over two sessions: first without CAD, then with either SE_CAD or SP_CAD, and after 1-month washout period, second session using the alternative CAD. Results: 375 patients (mean age, 65 ± 15 years; 170 women) were included. The dataset comprised 139 cases with pulmonary opacities (nodules / masses / consolidations), 119 with pleural effusion, 29 with pneumothorax, and 161 normal cases, including overlaps. SP_CAD-assisted readings showed higher specificity than SE_CAD-assisted for pulmonary opacities ( p < 0.001). Sensitivity did not differ significantly between SE_CAD and SP_CAD assistance. For pulmonary opacities, CAD-assisted reading showed higher sensitivity (both CAD: p < 0.001) but lower specificity (SE_CAD: p = 0.006; SP_CAD: p = 0.03) than unassisted reading. For pleural effusion, CAD-assisted improved sensitivity (SE_CAD: p < 0.001; SP_CAD: p = 0.02) and specificity (both CAD: p < 0.001) than unassisted reading. For pneumothorax, SE_CAD assistance did not significantly affect sensitivity but decreased specificity, whereas SP_CAD assistance improved sensitivity without significantly affecting specificity. Conclusion: The specificity-prioritized CAD showed higher specificity than the sensitivity-prioritized CAD, without a significant reduction in sensitivity. These findings suggest that different CAD operating points can influence radiologists’ diagnostic performance. Secondary We compared radiologist performance using sensitivity-optimized and specificity-optimized computer-aided detection (CAD). Different predefined CAD operating points influenced radiologist performance, resulting in differences in specificity for nodules/masses/consolidations. Our results show that different preset decision thresholds within the same AI model can influence diagnostic performance in AI-assisted radiograph interpretation.

Rationale: Conventional monitoring of acute respiratory distress syndrome (ARDS) relies on global parameters, eg, tidal volume, airway pressure, and driving pressure. These parameters do not capture regional stress heterogeneity within the lung. Objectives: To develop and validate a novel technique, lung stress mapping visualizing regional lung stress throughout the lung, and to evaluate its biological and clinical relevance. Methods: Lung stress mapping combines esophageal pressure and plateau pressure with CT-derived pleural pressure gradients to generate spatially resolved maps of inspiratory transpulmonary pressure. Accuracy was tested in pigs by surgically inserted pleural sensors. Biological relevance was assessed in rabbits by correlating lung stress mapping-derived parameters with regional proinflammatory cytokine expression. Clinical feasibility and associations with outcome were evaluated in 20 consecutive ARDS patients enrolled in a prospective study. Measurements and Main Results: Good correlation and agreement between sensor-derived and mapping-derived lung stress were confirmed. In rabbits, lung inflammation predominantly occurred in nondependent lung regions where lung stress was higher, and overall inflammation correlated with lung stress mapping-derived parameters. In ARDS patients, all received lung-protective ventilation. Non-survivors had significantly higher lung stress mapping-derived maximum and mean lung stress than survivors, despite similar global ventilatory parameters. Exploratory ROC analyses showed stronger associations of lung stress mapping-derived parameters with 90-day mortality than driving pressure. Conclusions: Lung stress mapping accurately quantified regional transpulmonary stress and revealed biologically and clinically meaningful heterogeneity. This technique may help identify patients with ARDS at increased risk of ventilator-induced lung injury who would not be recognized through conventional respiratory monitoring.

Background: To investigate the performance of radiologists in characterizing and diagnosing hepatic lesions with and without the assistance of deep learning-based artificial intelligence (AI). Methods: This retrospective study included 83 nodules/masses from 69 patients who underwent dynamic contrast-enhanced CT of the liver. Image assessments were conducted by 20 radiologists. grouped according to their level of experience (10 senior and 10 junior). Each radiologist determined the probability of eight characteristics based on enhancement patterns and the diagnosis with and without AI attached to the SYNAPSE SAI viewer (FUJIFILM Corporation, Minato-ku, Japan). The reference standard for comparison was established as follows: final diagnoses were based on pathology for 39 lesions and expert imaging consensus for the remainder, while image characteristics for all lesions were determined by expert imaging consensus. Areas under the receiver operating characteristic curves (AUCs) were analyzed using the multireader multicase method. Results: Using AI significantly improved the overall AUCs for both the characterization and the diagnosis of liver lesions. Improvement was suggested for specific items, including the characterization of enhancement, nonperipheral washout, and delayed enhancement, and the diagnosis of hepatocellular carcinoma. The utilization of AI system also suggested potential improvements in the AUCs for image characterization in both the senior and junior groups. Conclusions: Using AI improved the radiologists’ performance in characterizing and diagnosing hepatic lesions. In terms of their capacity to assess imaging characteristics, improvements were observed regardless of their level of experience.