Search

Contains fulltext : 305367.pdf (Publisher’s version ) (Open Access), The adoption of artificial intelligence (AI) tools in medicine poses challenges to existing clinical workflows. This commentary discusses the necessity of context-specific quality assurance (QA), emphasizing the need for robust QA measures with quality control (QC) procedures that encompass (1) acceptance testing (AT) before clinical use, (2) continuous QC monitoring, and (3) adequate user training. The discussion also covers essential components of AT and QA, illustrated with real-world examples. We also highlight what we see as the shared responsibility of manufacturers or vendors, regulators, healthcare systems, medical physicists, and clinicians to enact appropriate testing and oversight to ensure a safe and equitable transformation of medicine through AI.

01 februari 2023, Item does not contain fulltext, Rapid advances in artificial intelligence (AI) and machine learning, and specifically in deep learning (DL) techniques, have enabled broad application of these methods in health care. The promise of the DL approach has spurred further interest in computer-aided diagnosis (CAD) development and applications using both "traditional" machine learning methods and newer DL-based methods. We use the term CAD-AI to refer to this expanded clinical decision support environment that uses traditional and DL-based AI methods. Numerous studies have been published to date on the development of machine learning tools for computer-aided, or AI-assisted, clinical tasks. However, most of these machine learning models are not ready for clinical deployment. It is of paramount importance to ensure that a clinical decision support tool undergoes proper training and rigorous validation of its generalizability and robustness before adoption for patient care in the clinic. To address these important issues, the American Association of Physicists in Medicine (AAPM) Computer-Aided Image Analysis Subcommittee (CADSC) is charged, in part, to develop recommendations on practices and standards for the development and performance assessment of computer-aided decision support systems. The committee has previously published two opinion papers on the evaluation of CAD systems and issues associated with user training and quality assurance of these systems in the clinic. With machine learning techniques continuing to evolve and CAD applications expanding to new stages of the patient care process, the current task group report considers the broader issues common to the development of most, if not all, CAD-AI applications and their translation from the bench to the clinic. The goal is to bring attention to the proper training and validation of machine learning algorithms that may improve their generalizability and reliability and accelerate the adoption of CAD-AI systems for clinical decision support.

Background: Drug-induced interstitial lung disease (DIILD) is a form of interstitial lung disease resulting from exposure to drugs causing inflammation and possibly interstitial fibrosis. Antineoplastic drugs are the primary cause of DIILD, accounting for 23%-51% of cases, with bleomycin, everolimus, erlotinib, trastuzumab-deruxtecan and immune checkpoint inhibitors being the most common causative agents. DIILD can be difficult to identify and manage, and there are currently no specific guidelines on the diagnosis and treatment of DIILD caused by anticancer drugs.Objective: To develop recommendations for the diagnosis and management of DIILD in cancer patients.Methods: Based on the published literature and their clinical expertise, a multidisciplinary group of experts in Italy developed recommendations stratified by DIILD severity, based on the Common Terminology Criteria for Adverse Events.Results: The recommendations highlight the importance of multidisciplinary interaction in the diagnosis and management of DIILD. Important components of the diagnostic process are physical examination and careful patient history-taking, measurement of vital signs (particularly respiratory rate and arterial oxygen saturation), relevant laboratory tests, respiratory function testing with spirometry and diffusing capacity of the lung for carbon monoxide and computed tomography/imaging. Because the clinical and radiological signs of DIILD are often similar to those of pneumonias or interstitial lung diseases, differential diagnosis is important, including microbial and serological testing to exclude or confirm infectious causes. In most cases, management of DIILD requires the discontinuation of the antineoplastic agent and the administration of short-term steroids. Steroid tapering must be undertaken slowly to prevent reactivation of DIILD. Patients with severe and very severe (grade 3 and 4) DIILD will require hospitalisation and often need oxygen and non-invasive ventilation. Decis

Radiomics is a process of extraction and analysis of quantitative features from diagnostic images. Liquid biopsy is a test done on a sample of blood to look for cancer cells or for pieces of tumourigenic DNA circulating in the blood. Radiomics and liquid biopsy have great potential in oncology, since both are minimally invasive, easy to perform, and can be repeated in patient follow-up visits, enabling the extraction of valuable information regarding tumour type, aggressiveness, progression, and response to treatment. Both methods are in their infancy, with major evidence of application in lung and gastrointestinal cancer, while still undergoing evaluation in other cancer types. In this paper, the main oncologic applications of radiomics and liquid biopsy are reviewed, and a synergistic approach incorporating both tests for cancer diagnosis and follow-up is discussed within the context of systems medicine. Teaching Points: • Radiomics is a process of extraction and analysis of quantitative features from diagnostic images. • Most clinical applications of radiomics are in the field of oncologic imaging. • Radiomics applies to all imaging modalities. • A cluster of radiomic features is a “radiomic signature”. • Machine learning may improve the efficacy of radiomics analysis.

PURPOSE: To assess the effect of computer-aided detection (CAD) as a second reader on the sensitivity and specificity of computed tomographic (CT) colonography in detecting 6-9-mm colorectal cancer (CRC) lesions. MATERIALS AND METHODS: Individuals with clinical indications for colonoscopy--either for symptoms or as part of participating in a surveillance program or CRC screening--were prospectively enrolled at one of 10 academic centers between July 2007 and May 2009. Institutional review board approval was obtained at each clinical site, and all participants provided written informed consent. All participants underwent CT colonography and colonoscopy on the same day. Experienced readers interpreted the CT colonography images unassisted and then reviewed all colorectal lesion-like structures pinpointed by the CAD algorithm. Segmental unblinding of CT colonoscopy findings at colonoscopy was utilized. The sensitivity and specificity of unassisted and CAD-assisted reading in identifying individuals with 6-9-mm lesions were calculated and compared by means of pairwise analysis. RESULTS: A total of 618 participants (mean age, 57.9 years; 54.5% male) were included in the final analysis. Of these participants, 464 (75.1%) had no lesions 6 mm or larger, and 52 (8.4%) had 6-9-mm lesions. The sensitivity of CT colonography with unassisted reading and that with CAD-assisted reading in identifying individuals with 6-9-mm lesions was 65.4% (95% confidence interval [CI]: 50.9%, 78.0%) and 76.9% (95% CI: 63.2%, 87.5%; P = .016), respectively. No significant change in specificity was observed: The specificity of CT colonography with unassisted and that with CAD-assisted reading was 91.8% (95% CI: 88.9%, 94.1%) and 90.9% (95% CI: 88.0%, 93.4%; P = .063), respectively. Evaluation of CAD candidates required an additional 1.6 minutes (25th-75th percentile: 1.0 minute to 3.4 minutes). CONCLUSION: The addition of CAD to reading performed by experienced readers resulted in a significant b