Kataoka, Kensuke

ABSTRACTBackground and ObjectiveThe diagnosis of interstitial lung diseases (ILDs) often relies on the integration of various clinical, radiological, and histopathological data. Achieving high diagnostic accuracy in ILDs, particularly for distinguishing usual interstitial pneumonia (UIP), is challenging and requires a multidisciplinary approach. Therefore, this study aimed to develop a multimodal artificial intelligence (AI) algorithm that combines computed tomography (CT) and histopathological images to improve the accuracy and consistency of UIP diagnosis.MethodsA dataset of CT and pathological images from 324 patients with ILD between 2009 and 2021 was collected. The CT component of the model was trained to identify 28 different radiological features. The pathological counterpart was developed in our previous study. A total of 114 samples were selected and used for testing the multimodal AI model. The performance of the multimodal AI was assessed through comparisons with expert pathologists and general pathologists.ResultsThe developed multimodal AI demonstrated a substantial improvement in distinguishing UIP from non‐UIP, achieving an AUC of 0.92. When applied by general pathologists, the diagnostic agreement rate improved significantly, with a post‐model κ score of 0.737 compared to 0.273 pre‐model integration. Additionally, the diagnostic consensus rate with expert pulmonary pathologists increased from κ scores of 0.278–0.53 to 0.474–0.602 post‐model integration. The model also increased diagnostic confidence among general pathologists.ConclusionCombining CT and histopathological images, the multimodal AI algorithm enhances pathologists' diagnostic accuracy, consistency, and confidence in identifying UIP, even in cases where specialised expertise is limited. image

AbstractBackground and objectiveDespite significant recent progress in the understanding of idiopathic pulmonary fibrosis (IPF), the early phase of the disease is still poorly understood. We studied patients with IPF without pulmonary function impairment in order to determine the clinical features, natural history and key findings for physiological progression.MethodsTwenty‐five patients without pulmonary function impairment were identified from among patients with previously diagnosed IPF who underwent surgical lung biopsies between January 1997 and December 2006 at our institutions. ‘Without pulmonary function impairment’ was defined as both forced vital capacity (FVC) and diffusing capacity for carbon monoxide (DLco) >80% predicted. Patients diagnosed with IPF through multidisciplinary discussion based on the new IPF guidelines were the subjects of this study.ResultsSixteen patients had a confirmed diagnosis of IPF. Eleven patients presented with chest X‐ray abnormality found during an annual health examination. Seven patients were asymptomatic. Eleven patients showed physiological disease progression (median time; 19.9 ± 12.3 months) defined by a decline of at least 10% in FVC or at least 15% in DLco. Univariate analysis revealed that both usual interstitial pneumonia pattern and extent of honeycombing on high‐resolution computed tomography (HRCT) were factors associated with disease progression (odds ratio 5.634, 95% confidence interval 1.364–23.278; odds ratio 2.371/5%, 95% confidence interval 1.042–5.395).ConclusionsIPF patients without pulmonary function impairment have a progressive but slow clinical course. The existence and extent of honeycombing on HRCT are predictive of disease progression.