Deep learning algorithms for melanoma detection using dermoscopic images: A systematic review and meta-analysis.

BACKGROUND: Melanoma is a serious risk to human health and early identification is vital for treatment success. Deep learning (DL) has the potential to detect cancer using imaging technologies and many studies provide evidence that DL algorithms can achieve high accuracy in melanoma diagnostics. OBJECTIVES: To critically assess different DL performances in diagnosing melanoma using dermatoscopic images and discuss the relationship between dermatologists and DL. METHODS: Ovid-Medline, Embase, IEEE Xplore, and the Cochrane Library were systematically searched from inception until 7th December 2021. Studies that reported diagnostic DL model performances in detecting melanoma using dermatoscopic images were included if they had specific outcomes and histopathologic confirmation. Binary diagnostic accuracy data and contingency tables were extracted to analyze outcomes of interest, which included sensitivity (SEN), specificity (SPE), and area under the curve (AUC). Subgroup analyses were performed according to human-machine comparison and cooperation. The study was registered in PROSPERO, CRD42022367824. RESULTS: 2309 records were initially retrieved, of which 37 studies met our inclusion criteria, and 27 provided sufficient data for meta-analytical synthesis. The pooled SEN was 82 % (range 77-86), SPE was 87 % (range 84-90), with an AUC of 0.92 (range 0.89-0.94). Human-machine comparison had pooled AUCs of 0.87 (0.84-0.90) and 0.83 (0.79-0.86) for DL and dermatologists, respectively. Pooled AUCs were 0.90 (0.87-0.93), 0.80 (0.76-0.83), and 0.88 (0.85-0.91) for DL, and junior and senior dermatologists, respectively. Analyses of human-machine cooperation were 0.88 (0.85-0.91) for DL, 0.76 (0.72-0.79) for unassisted, and 0.87 (0.84-0.90) for DL-assisted dermatologists. CONCLUSIONS: Evidence suggests that DL algorithms are as accurate as senior dermatologists in melanoma diagnostics. Therefore, DL could be used to support dermatologists in diagnostic decision-making. Although, further high-quality, large-scale multicenter studies are required to address the specific challenges associated with medical AI-based diagnostics.

Distribution and diagnostic value of single and multiple high-risk HPV infections in detection of cervical intraepithelial neoplasia: A retrospective multicenter study in China.

The risk associated with single and multiple human papillomavirus (HPV) infections in cervical intraepithelial neoplasia (CIN) remains uncertain. This study aims to explore the distribution and diagnostic significance of the number of high-risk HPV (hr-HPV) infections in detecting CIN, addressing a crucial gap in our understanding. This comprehensive multicenter, retrospective study meticulously analyzed the distribution of single and multiple hr-HPV, the risk of CIN2+, the relationship with CIN, and the impact on the diagnostic performance of colposcopy using demographic information, clinical histories, and tissue samples. The composition of a single infection was predominantly HPV16, 52, 58, 18, and 51, while HPV16 and 33 were identified as the primary causes of CIN2+. The primary instances of dual infection were mainly observed in combinations such as HPV16/18, HPV16/52, and HPV16/58, while HPV16/33 was identified as the primary cause of CIN2+. The incidence of hr-HPV infections shows a dose-response relationship with the risk of CIN (p for trend <0.001). Compared to single hr-HPV, multiple hr-HPV infections were associated with increased risks of CIN1 (1.44, 95% confidence interval [CI]: 1.20-1.72), CIN2 (1.70, 95% CI: 1.38-2.09), and CIN3 (1.08, 95% CI: 0.86-1.37). The colposcopy-based specificity of single hr-HPV (93.4, 95% CI: 92.4-94.4) and multiple hr-HPV (92.9, 95% CI: 90.8-94.6) was significantly lower than negative (97.9, 95% CI: 97.0-98.5) in detecting high-grade squamous intraepithelial lesion or worse (HSIL+). However, the sensitivity of single hr-HPV (73.5, 95% CI: 70.8-76.0) and multiple hr-HPV (71.8, 95% CI: 67.0-76.2) was higher than negative (62.0, 95% CI: 51.0-71.9) in detecting HSIL+. We found that multiple hr-HPV infections increase the risk of developing CIN lesions compared to a single infection. Colposcopy for HSIL+ detection showed high sensitivity and low specificity for hr-HPV infection. Apart from HPV16, this study also found that HPV33 is a major pathogenic genotype.

The Causal Relationship between the Morning Chronotype and the Gut Microbiota: A Bidirectional Two-Sample Mendelian Randomization Study.

BACKGROUND: Numerous observational studies have documented an association between the circadian rhythm and the composition of the gut microbiota. However, the bidirectional causal effect of the morning chronotype on the gut microbiota is unknown. METHODS: A two-sample Mendelian randomization study was performed, using the summary statistics of the morning chronotype from the European Consortium and those of the gut microbiota from the largest available genome-wide association study meta-analysis, conducted by the MiBioGen consortium. The inverse variance-weighted (IVW), weighted mode, weighted median, MR-Egger regression, and simple mode methods were used to examine the causal association between the morning chronotype and the gut microbiota. A reverse Mendelian randomization analysis was conducted on the gut microbiota, which was identified as causally linked to the morning chronotype in the initial Mendelian randomization analysis. Cochran's Q statistics were employed to assess the heterogeneity of the instrumental variables. RESULTS: Inverse variance-weighted estimates suggested that the morning chronotype had a protective effect on Family Bacteroidaceae (ß = -0.072; 95% CI: -0.143, -0.001; p = 0.047), Genus Parabacteroides (ß = -0.112; 95% CI: -0.184, -0.039; p = 0.002), and Genus Bacteroides (ß = -0.072; 95% CI: -0.143, -0.001; p = 0.047). In addition, the gut microbiota (Family Bacteroidaceae (OR = 0.925; 95% CI: 0.857, 0.999; p = 0.047), Genus Parabacteroides (OR = 0.915; 95% CI: 0.858, 0.975; p = 0.007), and Genus Bacteroides (OR = 0.925; 95% CI: 0.857, 0.999; p = 0.047)) demonstrated positive effects on the morning chronotype. No significant heterogeneity in the instrumental variables, or in horizontal pleiotropy, was found. CONCLUSION: This two-sample Mendelian randomization study found that Family Bacteroidaceae, Genus Parabacteroides, and Genus Bacteroides were causally associated with the morning chronotype. Further randomized controlled trials are needed to clarify the effects of the gut microbiota on the morning chronotype, as well as their specific protective mechanisms.