Artificial intelligence for detection of microsatellite instability in colorectal cancer-a multicentric analysis of a pre-screening tool for clinical application.

BACKGROUND: Microsatellite instability (MSI)/mismatch repair deficiency (dMMR) is a key genetic feature which should be tested in every patient with colorectal cancer (CRC) according to medical guidelines. Artificial intelligence (AI) methods can detect MSI/dMMR directly in routine pathology slides, but the test performance has not been systematically investigated with predefined test thresholds. METHOD: We trained and validated AI-based MSI/dMMR detectors and evaluated predefined performance metrics using nine patient cohorts of 8343 patients across different countries and ethnicities. RESULTS: Classifiers achieved clinical-grade performance, yielding an area under the receiver operating curve (AUROC) of up to 0.96 without using any manual annotations. Subsequently, we show that the AI system can be applied as a rule-out test: by using cohort-specific thresholds, on average 52.73% of tumors in each surgical cohort [total number of MSI/dMMR = 1020, microsatellite stable (MSS)/ proficient mismatch repair (pMMR) = 7323 patients] could be identified as MSS/pMMR with a fixed sensitivity at 95%. In an additional cohort of N = 1530 (MSI/dMMR = 211, MSS/pMMR = 1319) endoscopy biopsy samples, the system achieved an AUROC of 0.89, and the cohort-specific threshold ruled out 44.12% of tumors with a fixed sensitivity at 95%. As a more robust alternative to cohort-specific thresholds, we showed that with a fixed threshold of 0.25 for all the cohorts, we can rule-out 25.51% in surgical specimens and 6.10% in biopsies. INTERPRETATION: When applied in a clinical setting, this means that the AI system can rule out MSI/dMMR in a quarter (with global thresholds) or half of all CRC patients (with local fine-tuning), thereby reducing cost and turnaround time for molecular profiling.

[Computer-assisted skin cancer diagnosis : Is it time for artificial intelligence in clinical practice?] / Computerassistierte Hautkrebsdiagnose : Wann kommt künstliche Intelligenz in der Praxis an?

BACKGROUND: Artificial intelligence (AI) is increasingly being used in medical practice. Especially in the image-based diagnosis of skin cancer, AI shows great potential. However, there is a significant discrepancy between expectations and true relevance of AI in current dermatological practice. OBJECTIVES: This article summarizes promising study results of skin cancer diagnosis by computer-based diagnostic systems and discusses their significance for daily practice. We hereby focus on the analysis of dermoscopic images of pigmented and unpigmented skin lesions. MATERIALS AND METHODS: A selective literature search for recent relevant trials was conducted. The included studies used machine learning, and in particular "convolutional neural networks", which have been shown to be particularly effective for the classification of image data. RESULTS AND CONCLUSIONS: In numerous studies, computer algorithms were able to detect pigmented and nonpigmented neoplasms of the skin with high precision, comparable to that of dermatologists. The combination of the physician's assessment and AI showed the best results. Computer-based diagnostic systems are widely accepted among patients and physicians. However, they are still not applicable in daily practice, since computer-based diagnostic systems have only been tested in an experimental environment. In addition, many digital diagnostic criteria that help AI to classify skin lesions remain unclear. This lack of transparency still needs to be addressed. Moreover, clinical studies on the use of AI-based assistance systems are needed in order to prove its applicability in daily dermatologic practice.

2-mm surgical margins are adequate for most basal cell carcinomas in Japanese: a retrospective multicentre study on 1000 basal cell carcinomas.

BACKGROUND: Surgery is the gold standard for basal cell carcinomas (BCC). Current recommended surgical margins for BCCs are determined from studies in Caucasian populations. However, the appropriate surgical margins for BCCs in non-white races are unclear. OBJECTIVES: To investigate the accuracy of preoperative determination of clinical tumour borders and appropriate surgical margins in Japanese patients with BCC. METHODS: The maximum calculated differences in distance between the preoperatively determined surgical margins and the actual histologic tumour side margins were considered as 'accuracy gaps' of clinical tumour borders. Estimated side margin positivity rates (ESMPRs) with narrower (2 and 3 mm) surgical margins were calculated on the basis of the accuracy gaps. RESULTS: Overall, 1000 surgically excised BCCs from 980 Japanese patients were included. The most frequent histologic subtype was nodular BCC (67%). The median accuracy gap was 0.3 mm [interquartile range (IQR): -0.5 to +1 mm]. The ESMPRs with 2- and 3-mm surgical margins were 3.8% and 1.4%, respectively. Only the ESMPRs between the well-defined (n = 921) and poorly defined clinical tumour border groups (n = 79) showed statistical difference [2-mm margin: 3.1% vs. 11.7%, OR: 3.89, 95% confidential interval (CI): 1.41-10.71, P <0.01; 3-mm margin: 0.97% vs. 6.3%, OR: 6.58, 95% CI: 1.67-25.99, P <0.01]. No significant differences in ESMPRs were noted in other subgroups including risk classifications. CONCLUSIONS: The determined clinical tumour border accuracy gaps in this Japanese cohort were negligible. Dermatologic surgeons may use narrower surgical margins with acceptable margin positivity rates. The clarity of clinical tumour borders could be an appropriate guide for selection of different surgical margins in the Japanese cohort.

Psoriasis, cardiovascular risk factors and metabolic disorders: sex-specific findings of a population-based study.

BACKGROUND: Scientific evidence suggests an association between psoriasis and cardiovascular and metabolic diseases. However, there are hardly any sex-specific results from population-based studies reporting the prevalence of cardiovascular risk factors in patients with psoriasis and point estimates of the association between psoriasis and cardiovascular and metabolic disorders. OBJECTIVE: Aims are to evaluate the sex-specific prevalence of psoriasis and cardiovascular risk factors, and to estimate sex-specific associations between psoriasis and diabetes type 2 (DM) and metabolic syndrome (MetS). METHODS: We used data of 3723 participants (45-75 years, 54.1% women) without coronary heart disease and missing data (psoriasis, DM, MetS) from the Heinz Nixdorf Recall study. Standardized information on health outcomes and risk factors was assessed. We performed descriptive statistics and multiple regression analyses to calculate prevalence rate ratios (PR) and 95% confidence intervals (95% CI). RESULTS: The prevalence of psoriasis was 3.8% (n = 143), with no differences between sex. We observed more often metabolic and cardiovascular risk factors in women with psoriasis compared to women without psoriasis. Interestingly, in men, this pattern was partly reversed. Multiple regression analyses revealed distinctly elevated PRs for DM for both women and men with psoriasis (fully adjusted PR: 2.43; 95% CI: 1.17-5.07, resp. 2.09; 1.16-3.76). Regarding the MetS, the results were inconsistent, showing a positive association between psoriasis and MetS in women (1.84; 1.14-2.98), but a negative association in men, even though with a wide 95% CI (0.69; 0.42-1.12). CONCLUSION: The results of our cross-sectional, population-based analysis show a distinct association between psoriasis and DM, whereas for the MetS the results contrasted between men and women, translating in women with MetS showing a higher and in men a lower chance to be psoriatic. Our results emphasize the urgent need for sex-specific research, studying the effects of psoriasis on metabolic disorders as well as effective sex tailored prevention measures.