Search

Background and Aim: False positives (FPs) pose a significant challenge in the application of artificial intelligence (AI) for polyp detection during colonoscopy. The study aimed to quantitatively evaluate the impact of computer‐aided polyp detection (CADe) systems' FPs on endoscopists. Methods: The model's FPs were categorized into four gradients: 0–5, 5–10, 10–15, and 15–20 FPs per minute (FPPM). Fifty‐six colonoscopy videos were collected for a crossover study involving 10 endoscopists. Polyp missed rate (PMR) was set as primary outcome. Subsequently, to further verify the impact of FPPM on the assistance capability of AI in clinical environments, a secondary analysis was conducted on a prospective randomized controlled trial (RCT) from Renmin Hospital of Wuhan University in China from July 1 to October 15, 2020, with the adenoma detection rate (ADR) as primary outcome. Results: Compared with routine group, CADe reduced PMR when FPPM was less than 5. However, with the continuous increase of FPPM, the beneficial effect of CADe gradually weakens. For secondary analysis of RCT, a total of 956 patients were enrolled. In AI‐assisted group, ADR is higher when FPPM ≤ 5 compared with FPPM > 5 (CADe group: 27.78% vs 11.90%; P = 0.014; odds ratio [OR], 0.351; 95% confidence interval [CI], 0.152–0.812; COMBO group: 38.40% vs 23.46%, P = 0.029; OR, 0.427; 95% CI, 0.199–0.916). After AI intervention, ADR increased when FPPM ≤ 5 (27.78% vs 14.76%; P = 0.001; OR, 0.399; 95% CI, 0.231–0.690), but no statistically significant difference was found when FPPM > 5 (11.90% vs 14.76%, P = 0.788; OR, 1.111; 95% CI, 0.514–2.403). Conclusion: The level of FPs of CADe does affect its effectiveness as an aid to endoscopists, with its best effect when FPPM is less than 5. [ABSTRACT FROM AUTHOR]

Objectives: Cervical cancer is the most common malignant tumors in women leading to serious morbidity and mortality worldwide, especially among developing countries. A main cause of the disease is the high-risk human papillomavirus (HR-HPV) infection. HSIL usually progress to cervical cancer, and low-grade lesions, including LSIL and ASCUS, mostly turn to normal or benign lesions, but there are still a small number of patients who will progress to HSIL. Up to now there is no efficient biomarker clinically available to predict people with high risk to progress into HSIL. This study was conducted to evaluate the value of human papillomavirus (HPV) DNA, p16INK4a protein, and HPV L1 capsid protein in predicting HSIL and minimizing unnecessary colposcopy treatments.Methods: 1222 patients with HR-HPV infection or with abnormal Thinprep cytologic test (TCT) were chosen to conduct colposcopy in the cervical out-patient clinic of Shanghai First Maternity and Infant Hospital affiliated to Shanghai Tongji University from June 2014 to January 2017. TCT, cervical biopsy, HPV DNA and HPVL1 were performed on all patients. 110 patients were selected to detect p16INK4a protein. Hybrid capture 2 (HC-2) was used to detect HPV DNA, and their subgroups using gene typing system. Immunohistochemical technology was used to detect HPV L1 and p16.Results: HPV DNA was positive in 1097 cases, with the positive rate of 89.7% (1097/1222). In particular, the positive expression rates of HPV DNA were 82.3, 95.7, 96.6 and 100% in Normal/CC, LSIL, HSIL and cervical cancer groups, respectively (p < 0.001). HPV L1 was negative in 781 cases with HR-HPV infection, and the overall negative rate is 71.1%. In patients with Normal/CC, LSIL and HSIL, the negative expression rates of HPV L1 were 91.3, 40 and 81.2%, respectively (p value < 0.001). In the 110 patients, HPV L1 was negative in 98.1% (53/54) of Normal/CC, 42.9% (12/28) of LSIL and 85.1% (23/27) of HSIL (p value = 0.0043). P16-positive rates in patients with Normal/CC, LSIL and HSIL were 33.3% (18/54), 75% (21/28) and 96.2% (26/27), respectively (p value < 0.001). 18 out of 28 cases express low positive (+) in LSIL, 25 out of 27 cases express strong positive (3+) in HSIL. Patients with L1(-) p16(+) including 18.5% (10/54) of normal/cervicitis, 60.7% (17/28) of LSIL and 85.1% (23/27) of HSIL (p value < 0.005). Furthermore, patients with L1(-) p16(1+) included 37% (10/27) of normal/cervicitis 59.3% (16/27) of LSIL and 3.7% (1/27) of HSIL; patients with L1(-) p16(2+) consisted of 0% of normal/cervicitis/LSIL and 100% (1/1) of HSIL; patients with L1(-) p16(3+) were composed of 0% of normal/cervicitis, 4.5% (1/22) of LSIL and 95.5% (21/22) of HSIL (p value < 0.005) (Table 6).Conclusion: With the increase in the degree of the cervical lesions, the expression of HPV DNA and p16 is up-regulated while HPV L1 protein is down-regulated. HPV DNA, HPV L1 and p16 are useful markers for the prediction of HSIL. Combined detection of these three markers has important potential to predicting HSIL and minimizing unnecessary colposcope examination. [ABSTRACT FROM AUTHOR]

Persistent human papillomavirus (HPV) infection is the main causative agent for cervical intraepithelial neoplasia (CIN) and cancer. Variability in host immunogenetic factors is important in determining the overall cellular immune response to the HPV infection. This study was carried out to confirm the association of human leukocyte antigen (HLA) class II DRB1 and DQB1 alleles with CIN and HPV persistent infections in women from Shanghai in a case-controlled study. A total of 170 patients, including 105 HPV positive patients and 65 HPV negative women (control) participated in the study. HybriBio's proprietary flow-through hybridization technique was used to perform HPV genotyping. Low-resolution PCR-sequence specific priming (PCR-SSP) was used to genotype HLA class II for DRB1 and DQB1 loci. Binary and multivariate logistic regression analysis highlighted the association of specific alleles with CIN and HPV persistent infections after adjusting for the confounding factor of age. HLA-DQB1*02 and *06 is significantly associated with increased risk of HPV16 persistent infection ( P c < 0.013). HLA-DRB1*09 is significantly associated with increased risk for CIN, whereas the -DRB1*16 exhibit protective to CIN ( P < 0.05). Significant association is found for HLA-DQB1*04 and *06 with increased risk for CIN ( P < 0.05). There were possible associations of specific HLA class II alleles either with risk of persistent HPV infection or with developing CIN., Competing Interests: None., (IJCEP Copyright © 2018.)