Constructing a clinical radiographic knee osteoarthritis database using artificial intelligence tools with limited human labor: A proof of principle.

OBJECTIVE: To create a scalable and feasible retrospective consecutive knee osteoarthritis (OA) radiographic database with limited human labor using commercial and custom-built artificial intelligence (AI) tools. METHODS: We applied four AI tools, two commercially available and two custom-built tools, to analyze 6 years of clinical consecutive knee radiographs from patients aged 35-79 at the University of Copenhagen Hospital, Bispebjerg-Frederiksberg Hospital, Denmark. The tools provided Kellgren-Lawrence (KL) grades, joint space widths, patella osteophyte detection, radiographic view detection, knee joint implant detection, and radiographic marker detection. RESULTS: In total, 25,778 knee radiographs from 8575 patients were included in the database after excluding inapplicable radiographs, and 92.5% of the knees had a complete OA dataset. Using the four AI tools, we saved about 800 hours of radiologist reading time and only manually reviewed 16.0% of the images in the database. CONCLUSIONS: This study shows that clinical knee OA databases can be built using AI with limited human reading time for uniform grading and measurements. The concept is scalable temporally and across geographic regions and could help diversify further OA research by efficiently including radiographic knee OA data from different populations globally. We can prevent data dredging and overfitting OA theories on existing trite cohorts by including various gene pools and continuous expansion of new clinical cohorts. Furthermore, the suggested tools and applied approaches provide an ability to retest previous hypotheses and test new hypotheses on real-life clinical data with current disease prevalence and trends.

Diagnostic test accuracy study of a commercially available deep learning algorithm for ischemic lesion detection on brain MRIs in suspected stroke patients from a non-comprehensive stroke center.

PURPOSE: To estimate the ability of a commercially available artificial intelligence (AI) tool to detect acute brain ischemia on Magnetic Resonance Imaging (MRI), compared to an experienced neuroradiologist. METHODS: We retrospectively included 1030 patients with brain MRI, suspected of stroke from January 6th, 2020 to 1st of April 2022, based on these criteria: Age ≥ 18 years, symptoms within four weeks before the scan. The neuroradiologist reinterpreted the MRI scans and subclassified ischemic lesions for reference. We excluded scans with interpretation difficulties due to artifacts or missing sequences. Four MRI scanner models from the same vendor were used. The first 800 patients were included consecutively, remaining enriched for less frequent lesions. The index test was a CE-approved AI tool (Apollo version 2.1.1 by Cerebriu). RESULTS: The final analysis cohort comprised 995 patients (mean age 69 years, 53 % female). A case-based analysis for detecting acute ischemic lesions showed a sensitivity of 89 % (95 % CI: 85 %-91 %) and specificity of 90 % (95 % CI: 87 %-92 %). We found no significant difference in sensitivity or specificity based on sex, age, or comorbidities. Specificity was reduced in cases with DWI artifacts. Multivariate analysis showed that increasing ischemic lesion size and fragmented lesions were independently associated with higher sensitivity, while non-acute lesion ages lowered sensitivity. CONCLUSIONS: The AI tool exhibits high sensitivity and specificity in detecting acute ischemic lesions on MRI compared to an experienced neuroradiologist. While sensitivity depends on the ischemic lesions' characteristics, specificity depends on the image quality.

Should artificial intelligence have lower acceptable error rates than humans?

The first patient was misclassified in the diagnostic conclusion according to a local clinical expert opinion in a new clinical implementation of a knee osteoarthritis artificial intelligence (AI) algorithm at Bispebjerg-Frederiksberg University Hospital, Copenhagen, Denmark. In preparation for the evaluation of the AI algorithm, the implementation team collaborated with internal and external partners to plan workflows, and the algorithm was externally validated. After the misclassification, the team was left wondering: what is an acceptable error rate for a low-risk AI diagnostic algorithm? A survey among employees at the Department of Radiology showed significantly lower acceptable error rates for AI (6.8 %) than humans (11.3 %). A general mistrust of AI could cause the discrepancy in acceptable errors. AI may have the disadvantage of limited social capital and likeability compared to human co-workers, and therefore, less potential for forgiveness. Future AI development and implementation require further investigation of the fear of AI's unknown errors to enhance the trustworthiness of perceiving AI as a co-worker. Benchmark tools, transparency, and explainability are also needed to evaluate AI algorithms in clinical implementations to ensure acceptable performance.

Impact of Concurrent Use of Artificial Intelligence Tools on Radiologists Reading Time: A Prospective Feasibility Study.

RATIONAL AND OBJECTIVES: This study investigated how an AI tool impacted radiologists reading time for non-contrast chest CT exams. MATERIALS AND METHODS: An AI tool was implemented into the PACS reading workflow of non-contrast chest CT exams between April and May 2020. The reading time was recorded for one CONSULTANT RADIOLOGIST and one RADIOLOGY RESIDENT by an external observer. After each case radiologists answered questions regarding additional findings and perceived case overview. Reading times were recorded for 25 cases without and 20 cases with AI tool assistance for each reader. Differences in reading time with and without the AI tool were assessed using Welch's t-test for non-inferiority with non-inferiority limits defined as 100 seconds for the consultant and 200 seconds for the resident. RESULTS: The mean reading time for the radiology resident was not significantly affected by the AI tool (without AI 370s vs with AI 437s; +67s 95% CI -28s to +163s, p = 0.16). The reading time for the radiology consultant was also not significantly affected by the AI tool (without AI 366s vs with AI 380s; +13s (95% CI - -57s to 84s, p = 0.70). The AI tool led to additional actionable findings in 5/40 (12.5%) studies and better overview in 18/20 (90%) of studies for the resident. CONCLUSION: A PACS based implementation of an AI tool for concurrent reading of chest CT exams did not increase reading time with additional actionable findings made as well as a perceived better case overview for the radiology resident.

Utility of multiple rule out CT screening of high-risk atraumatic patients in an emergency department-a feasibility study.

BACKGROUND: Several large trials have evaluated the effect of CT screening based on specific symptoms, with varying outcomes. Screening of patients with CT based on their prognosis alone has not been examined before. For moderate-to-high risk patients presenting in the emergency department (ED), the potential gain from a CT scan might outweigh the risk of radiation exposure. We hypothesized that an accelerated "multiple rule out" CT screening of moderate-to-high risk patients will detect many clinically unrecognized diagnoses that affect change in treatment. METHOD: Patients ≥ 40 years, triaged as high-risk or moderate-to-high risk according to vital signs, were eligible for inclusion. Patients were scanned with a combined ECG-gated and dual energy CT scan of cerebrum, thorax, and abdomen. The impact of the CT scan on patient diagnosis and treatment was examined prospectively by an expert panel. RESULTS: A total of 100 patients were included in the study, (53% female, mean age 73 years [age range, 43-93]). The scan lead to change in treatment or additional examinations in 37 (37%) patients, of which 24 (24%) were diagnostically significant, change in acute treatment in 11 (11%) cases and previously unrecognized malignant tumors in 10 (10%) cases. The mean size specific radiation dose was 15.9 mSv (± 3.1 mSv). CONCLUSION: Screening with a multi-rule out CT scan of high-risk patients in an ED is feasible and result in discovery of clinically unrecognized diagnoses and malignant tumors, but at the cost of radiation exposure and downstream examinations. The clinical impact of these findings should be evaluated in a larger randomized cohort.