Browsing by Author "Wallace, Michael"

Now showing 1 - 4 of 4
  • Thumbnail Image
    Item
    Optimal Management of Malignant Polyps, From Endoscopic Assessment and Resection to Decisions About Surgery
    (Elsevier, 2018) Rex, Douglas K.; Shaukat, Aasma; Wallace, Michael; Medicine, School of Medicine
  • Thumbnail Image
    Item
    Performance of artificial intelligence for colonoscopy regarding adenoma and polyp detection: a meta-analysis
    (Elsevier, 2020) Hassan, Cesare; Spadaccini, Marco; Iannone, Andrea; Maselli, Roberta; Jovani, Manol; Chandrasekar, Viveksandeep Thoguluva; Antonelli, Giulio; Yu, Honggang; Areia, Miguel; Dinis-Ribeiro, Mario; Bhandari, Pradeep; Sharma, Prateek; Rex, Douglas K.; Rösch, Thomas; Wallace, Michael; Repici, Alessandro; Medicine, School of Medicine
    BACKGROUND AND AIMS One fourth of colorectal neoplasia is missed at screening colonoscopy, representing the main cause of interval colorectal cancer (CRC). Deep learning systems with real-time computer-aided polyp detection (CADe) showed high accuracy in artificial settings, and preliminary randomized clinical trials (RCT) reported favourable outcomes in clinical setting. Aim of this meta-analysis was to summarise available RCTs on the performance of CADe systems in colorectal neoplasia detection. METHODS We searched MEDLINE, EMBASE and Cochrane Central databases until March 2020 for RCTs reporting diagnostic accuracy of CADe systems in detection of colorectal neoplasia. Primary outcome was pooled adenoma detection rate (ADR), Secondary outcomes were adenoma per colonoscopy (APC) according to size, morphology and location, advanced APC (AAPC), as well as polyp detection rate (PDR), Polyp-per-colonoscopy (PPC), and sessile serrated lesion per colonoscopy (SPC). We calculated risk ratios (RR), performed subgroup, and sensitivity analysis, assessed heterogeneity, and publication bias. RESULTS Overall, 5 randomized controlled trials (4354 patients), were included in the final analysis. Pooled ADR was significantly higher in the CADe groups than in the control group (791/2163, 36.6% vs 558/2191, 25.2%; RR, 1.44; 95% CI, 1.27-1.62; p<0.01; I 2:42%). APC was also higher in the CADe group compared with control (1249/2163, 0.58 vs 779/2191, 0.36; RR, 1.70; 95% CI, 1.53-1.89; p<0.01;I 2:33%). APC was higher for <5 mm (RR, 1.69; 95% CI, 1.48-1.84), 6-9 mm (RR, 1.44; 95% CI, 1.19-1.75), and >10 mm adenomas (RR, 1.46; 95% CI, 1.04-2.06), as well as for proximal (RR, 1.59; 95% CI, 1.34-1.88) and distal (RR, 1.68; 95% CI, 1.50-1.88), and for flat (RR: 1.78 95% CI 1.47-2.15) and polypoid morphology (RR, 1.54; 95% CI, 1.40-1.68). Regarding histology, CADe resulted in a higher SPC (RR, 1.52; 95% CI,1.14-2.02), whereas a nonsignificant trend for AADR was found (RR, 1.35; 95% CI, 0.74 – 2.47; p = 0.33; I 2:69%). Level of evidence for RCTs was graded moderate. CONCLUSIONS According to available evidence, the incorporation of Artificial Intelligence as aid for detection of colorectal neoplasia results in a significant increase of the detection of colorectal neoplasia, and such effect is independent from main adenoma characteristics.
  • Thumbnail Image
    Item
    Radiomics Boosts Deep Learning Model for IPMN Classification
    (Springer, 2023) Yao, Lanhong; Zhang, Zheyuan; Demir, Ugur; Keles, Elif; Vendrami, Camila; Agarunov, Emil; Bolan, Candice; Schoots, Ivo; Bruno, Marc; Keswani, Rajesh; Miller, Frank; Gonda, Tamas; Yazici, Cemal; Tirkes, Temel; Wallace, Michael; Spampinato, Concetto; Bagci, Ulas; Radiology and Imaging Sciences, School of Medicine
    Intraductal Papillary Mucinous Neoplasm (IPMN) cysts are pre-malignant pancreas lesions, and they can progress into pancreatic cancer. Therefore, detecting and stratifying their risk level is of ultimate importance for effective treatment planning and disease control. However, this is a highly challenging task because of the diverse and irregular shape, texture, and size of the IPMN cysts as well as the pancreas. In this study, we propose a novel computer-aided diagnosis pipeline for IPMN risk classification from multi-contrast MRI scans. Our proposed analysis framework includes an efficient volumetric self-adapting segmentation strategy for pancreas delineation, followed by a newly designed deep learning-based classification scheme with a radiomics-based predictive approach. We test our proposed decision-fusion model in multi-center data sets of 246 multi-contrast MRI scans and obtain superior performance to the state of the art (SOTA) in this field. Our ablation studies demonstrate the significance of both radiomics and deep learning modules for achieving the new SOTA performance compared to international guidelines and published studies (81.9% vs 61.3% in accuracy). Our findings have important implications for clinical decision-making. In a series of rigorous experiments on multi-center data sets (246 MRI scans from five centers), we achieved unprecedented performance (81.9% accuracy). The code is available upon publication.
  • Thumbnail Image
    Item
    Right-Sided Location Not Associated With Missed Colorectal Adenomas in an Individual-Level Reanalysis of Tandem Colonoscopy Studies
    (Elsevier, 2019) Zimmermann-Fraedrich, Katharina; Sehner, Susanne; Rex, Douglas K.; Kaltenbach, Tonya; Soetniko, Roy; Wallace, Michael; Leung, Wai K.; Guo, Chuanguo; Gralnek, Ian M.; Brand, Eelco C.; Groth, Stefan; Schachschal, Guido; Ikematsu, Hiroaki; Siersema, Peter D.; Rösch, Thomas; Medicine, School of Medicine
    Background & Aims Interval cancers occur more frequently in the right colon. One reason could be that right-sided adenomas are frequently missed in colonoscopy examinations. We reanalyzed data from tandem colonoscopies to assess adenoma miss rates in relation to location and other factors. Methods We pooled data from 8 randomized tandem trials comprising 2218 patients who had diagnostic or screening colonoscopies (adenomas detected in 49.8% of patients). We performed a mixed-effects logistic regression with patients as cluster effects with different independent parameters. Factors analyzed included location (left vs right, splenic flexure as cutoff), adenoma size, form, and histologic features. Analyses were controlled for potential confounding factors such as patient sex and age, colonoscopy indication, and bowel cleanliness. Results Right-side location was not an independent risk factor for missed adenomas (odds ratio [OR] compared with the left side, 0.94; 95% CI, 0.75–1.17). However, compared with adenomas ≤5 mm, the OR for missing adenomas of 6–9 mm was 0.62 (95% CI, 0.44–0.87), and the OR for missing adenomas of ≥10 mm was 0.51 (95% CI, 0.33–0.77). Compared with pedunculated adenomas, sessile (OR, 1.82; 95% CI, 1.16–2.85) and flat adenomas (OR, 2.47; 95% CI, 1.49–4.10) were more likely to be missed. Histologic features were not significant risk factors for missed adenomas (OR for adenomas with high-grade intraepithelial neoplasia, 0.68; 95% CI, 0.34–1.37 and OR for sessile serrated adenomas, 0.87; 95% CI, 0.47–1.64 compared with low-grade adenomas). Men had a higher number of adenomas per colonoscopy (1.27; 95% CI, 1.21–1.33) than women (0.86; 95% CI, 0.80–0.93). Men were less likely to have missed adenomas than women (OR for missed adenomas in men, 0.73; 95% CI, 0.57–0.94). Conclusions In an analysis of data from 8 randomized trials, we found that right-side location of an adenoma does not increase its odds for being missed during colonoscopy but that adenoma size and histologic features do increase risk. Further studies are needed to determine why adenomas are more frequently missed during colonoscopies in women than men.