The objective of this study was to assess prospectively the diagnostic accuracy of computer-assisted computed tomographic colonography (CTC) in the detection of polypoid (pedunculated or sessile) and nonpolypoid neoplasms and compare the accuracy between gastroenterologists and radiologists. This nationwide multicenter prospective controlled trial recruited 1,257 participants with average or high risk of colorectal cancer at 14 Japanese institutions. Participants had CTC and colonoscopy on the same day. CTC images were interpreted independently by trained gastroenterologists and radiologists. The main outcome was the accuracy of CTC in the detection of neoplasms ≥6 mm in diameter, with colonoscopy results as the reference standard. Detection sensitivities of polypoid vs. nonpolypoid lesions were also evaluated. Of the 1,257 participants, 1,177 were included in the final analysis: 42 (3.6%) were at average risk of colorectal cancer, 456 (38.7%) were at elevated risk, and 679 (57.7%) had recent positive immunochemical fecal occult blood tests. The overall per-participant sensitivity, specificity, and positive and negative predictive values for neoplasms ≥6 mm in diameter were 0.90, 0.93, 0.83, and 0.96, respectively, among gastroenterologists and 0.86, 0.90, 0.76, and 0.95 among radiologists (P<0.05 for gastroenterologists vs. radiologists). The sensitivity and specificity for neoplasms ≥10 mm in diameter were 0.93 and 0.99 among gastroenterologists and 0.91 and 0.98 among radiologists (not significant for gastroenterologists vs. radiologists). The CTC interpretation time by radiologists was shorter than that by gastroenterologists (9.97 vs. 15.8 min, P<0.05). Sensitivities for pedunculated and sessile lesions exceeded those for flat elevated lesions ≥10 mm in diameter in both groups (gastroenterologists 0.95, 0.92, and 0.68; radiologists: 0.94, 0.87, and 0.61; P0.05) for gastroenterologists vs. radiologists. CTC interpretation by gastroenterologists and radiologists was accurate for detection of polypoid neoplasms, but less so for nonpolypoid neoplasms. Gastroenterologists had a higher accuracy in the detection of neoplasms ≥6 mm than did radiologists, although their interpretation time was longer than that of radiologists.

Innovation in medical imaging artificial intelligence (AI)/machine learning (ML) demands extensive data collection, algorithmic advancements, and rigorous performance assessments encompassing aspects such as generalizability, uncertainty, bias, fairness, trustworthiness, and interpretability. Achieving widespread integration of AI/ML algorithms into diverse clinical tasks will demand a steadfast commitment to overcoming issues in model design, development, and performance assessment. The complexities of AI/ML clinical translation present substantial challenges, requiring engagement with relevant stakeholders, assessment of cost-effectiveness for user and patient benefit, timely dissemination of information relevant to robust functioning throughout the AI/ML lifecycle, consideration of regulatory compliance, and feedback loops for real-world performance evidence. This commentary addresses several hurdles for the development and adoption of AI/ML technologies in medical imaging. Comprehensive attention to these underlying and often subtle factors is critical not only for tackling the challenges but also for exploring novel opportunities for the advancement of AI in radiology.

Principal composite kernel feature analysis (PC-KFA) is presented to show kernel adaptations for nonlinear features of medical image data sets (MIDS) in computer-aided diagnosis (CAD). The proposed algorithm PC-KFA has extended the existing studies on kernel feature analysis (KFA), which extracts salient features from a sample of unclassified patterns by use of a kernel method. The principal composite process for PC-KFA herein has been applied to kernel principal component analysis [34] and to our previously developed accelerated kernel feature analysis [20]. Unlike other kernel-based feature selection algorithms, PC-KFA iteratively constructs a linear subspace of a high-dimensional feature space by maximizing a variance condition for the nonlinearly transformed samples, which we call data-dependent kernel approach. The resulting kernel subspace can be first chosen by principal component analysis, and then be processed for composite kernel subspace through the efficient combination representations used for further reconstruction and classification. Numerical experiments based on several MID feature spaces of cancer CAD data have shown that PC-KFA generates efficient and an effective feature representation, and has yielded a better classification performance for the proposed composite kernel subspace using a simple pattern classifier.

Because of the rapid spread and wide range of the clinical manifestations of the coronavirus disease 2019 (COVID-19), fast and accurate estimation of the disease progression and mortality is vital for the management of the patients. Currently available image-based prognostic predictors for patients with COVID-19 are largely limited to semi-automated schemes with manually designed features and supervised learning, and the survival analysis is largely limited to logistic regression. We developed a weakly unsupervised conditional generative adversarial network, called pix2surv, which can be trained to estimate the time-to-event information for survival analysis directly from the chest computed tomography (CT) images of a patient. We show that the performance of pix2surv based on CT images significantly outperforms those of existing laboratory tests and image-based visual and quantitative predictors in estimating the disease progression and mortality of COVID-19 patients. Thus, pix2surv is a promising approach for performing image-based prognostic predictions.