Computer-assisted diagnosis for an early identification of lung cancer in chest X rays

Abstract

Computer-assisted diagnosis (CAD) algorithms have shown its usefulness for the identification of pulmonary nodules in chest x-rays, but its capability to diagnose lung cancer (LC) is unknown. A CAD algorithm for the identification of pulmonary nodules was created and used on a retrospective cohort of patients with x-rays performed in 2008 and not examined by a radiologist when obtained. X-rays were sorted according to the probability of pulmonary nodule, read by a radiologist and the evolution for the following three years was assessed. The CAD algorithm sorted 20,303 x-rays and defined four subgroups with 250 images each (percentiles ≥ 98, 66, 33 and 0). Fifty-eight pulmonary nodules were identified in the ≥ 98 percentile (23,2%), while only 64 were found in lower percentiles (8,5%) (p < 0.001). A pulmonary nodule was confirmed by the radiologist in 39 out of 173 patients in the high-probability group who had follow-up information (22.5%), and in 5 of them a LC was diagnosed with a delay of 11 months (12.8%). In one quarter of the chest x-rays considered as high-probability for pulmonary nodule by a CAD algorithm, the finding is confirmed and corresponds to an undiagnosed LC in one tenth of the cases.

Integrated Dataset-Preparation System for ML-Based Medical Image Diagnosis with High Clinical Applicability in Various Modalities and Diagnoses

“…Machine learning-based computer-aided diagnosis (MLbased CAD) is a field that involves analyzing large datasets of patient data, particularly medical images, to assist clinicians in decision-making. Numerous studies in this field have been conducted on different subjects and image modalities such as the characterization of breast tumors with MRI scans [1][2][3], the detection of cerebral aneurysms with CT angiographies [4,5], and the detection of lung nodules with chest X-rays [6,7]. Within such studies, the dataset serves as the foundation upon which ML models are trained, directly influencing performance of a CAD system.…”