Microsatellite instability (MSI) is a molecular marker of deficient DNA mismatch repair (dMMR) that is found in approximately 15% of colorectal cancer (CRC) patients. Testing all CRC patients for MSI/dMMR is recommended as screening for Lynch Syndrome and, more recently, to determine eligibility for immune checkpoint inhibitors in advanced disease. However, universal testing for MSI/dMMR has not been uniformly implemented because of cost and resource limitations. Artificial intelligence has been used to predict MSI/dMMR directly from hematoxylin and eosin (H&E) stained tissue slides. We review the emerging data regarding the utility of machine learning for MSI classification, focusing on CRC. We also provide the clinician with an introduction to image analysis with machine learning and convolutional neural networks. Machine learning can predict MSI/dMMR with high accuracy in high quality, curated datasets. Accuracy can be significantly decreased when applied to cohorts with different ethnic and/or clinical characteristics, or different tissue preparation protocols. Research is ongoing to determine the optimal machine learning methods for predicting MSI, which will need to be compared to current clinical practices, including next-generation sequencing. Predicting response to immunotherapy remains an unmet need.
Introduction: Understanding the trend of student authorship is crucial in determining its correlation to scholarly impact for corresponding authors. Our objective is to investigate student authorship rates over time in articles published in JAMA Internal Medicine (IM), as well as to examine potential effects student authors have on scholarly impact scores of corresponding authors via H-index measures. Methods: Authorship data including student authors (SA), first student authors, and corresponding authors (CA) from prior JAMA IM publications between 2010 and 2018 were collected, with a total of 701 studies. Analysis of variance (ANOVA) and independent sample t-tests were performed to assess for differences in the mean by publishing year and student authorship, respectively. Results: Of 4591 total authors, 683 (14.9%) were considered student authors. The percentage of student authorship increased from 46.3% to 58.0% between 2010 and 2018, respectively. No difference in average H-indices of CA between SA and non-SA groups (overall NSA H i mean: 30.2, vs SA H i mean: 32.1, p=0.371) was noted. Discussion: Student participation in research is not a disadvantage to scholarly impact for corresponding authors. Increased student authorship reflects a promising trend towards greater student participation in research within the field of medicine.
Background Routine preoperative staging in pancreas cancer is controversial. We sought to evaluate the rates of diagnostic laparoscopy (DLAP) for pancreatic cancer. Methods We queried the National Surgical Quality Improvement Program for patients with pancreas cancer (2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013) and compared groups who underwent DLAP, exploratory laparotomy (XLAP), pancreas resection (RSXN) or therapeutic bypass (THBP). We compared demographics, comorbidities, postoperative complications, 30-day mortality (Chi-square P \ 0.05) and trends over time (R 2 0-1). Results We identified 17,138 patients (RSXN 81.8%, XLAP 16.5%, THBP 8.2%, and DLAP 12.9%), with some having multiple CPT codes. Only 10.3% (n = 1432) of RSXN patients underwent DLAP prior to resection. XLAP occurred in 49.5% of non-RSXN patients, of whom 67.1% had no other operation. The percentage of patients undergoing RSXN increased 20.3% over time (R 2 0.81), while DLAP decreased 52.6% (R 2 0.92). XLAP patients without other operations decreased from 4.2 to 2.4%, although not linearly (R 2 0.31). Only 10.3% of XLAP had a diagnostic laparoscopy as well, leaving nearly 90% of these patients with an exploratory laparotomy without RSXN or THBP. Discussion Diagnostic laparoscopy for pancreas malignancy is becoming less common but could benefit a subset of patients who undergo open exploration without resection or therapeutic bypass.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.