Performance of Artificial Intelligence Content Detectors Using Human and Artificial Intelligence-Generated Scientific Writing.

BACKGROUND: Few studies have examined the performance of artificial intelligence (AI) content detection in scientific writing. This study evaluates the performance of publicly available AI content detectors when applied to both human-written and AI-generated scientific articles. METHODS: Articles published in Annals of Surgical Oncology (ASO) during the year 2022, as well as AI-generated articles using OpenAI's ChatGPT, were analyzed by three AI content detectors to assess the probability of AI-generated content. Full manuscripts and their individual sections were evaluated. Group comparisons and trend analyses were conducted by using ANOVA and linear regression. Classification performance was determined using area under the curve (AUC). RESULTS: A total of 449 original articles met inclusion criteria and were evaluated to determine the likelihood of being generated by AI. Each detector also evaluated 47 AI-generated articles by using titles from ASO articles. Human-written articles had an average probability of being AI-generated of 9.4% with significant differences between the detectors. Only two (0.4%) human-written manuscripts were detected as having a 0% probability of being AI-generated by all three detectors. Completely AI-generated articles were evaluated to have a higher average probability of being AI-generated (43.5%) with a range from 12.0 to 99.9%. CONCLUSIONS: This study demonstrates differences in the performance of various AI content detectors with the potential to label human-written articles as AI-generated. Any effort toward implementing AI detectors must include a strategy for continuous evaluation and validation as AI models and detectors rapidly evolve.

Preoperative SARS-CoV-2 infection increases risk of early postoperative cardiovascular complications following noncardiac surgery.

As the coronavirus disease 2019 (COVID-19) pandemic progresses to an endemic phase, a greater number of patients with a history of COVID-19 will undergo surgery. Major adverse cardiovascular and cerebrovascular events (MACE) are the primary contributors to postoperative morbidity and mortality; however, studies assessing the relationship between a previous severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection and postoperative MACE outcomes are limited. Here, we analyzed retrospective data from 457,804 patients within the N3C Data Enclave, the largest national, multi-institutional data set on COVID-19 in the United States. However, 7.4% of patients had a history of COVID-19 before surgery. When comorbidities, age, race, and risk of surgery were controlled, patients with preoperative COVID-19 had an increased risk for 30-day postoperative MACE. MACE risk was influenced by an interplay between COVID-19 disease severity and time between surgery and infection; in those with mild disease, MACE risk was not increased even among those undergoing surgery within 4 wk following infection. In those with moderate disease, risk for postoperative MACE was mitigated 8 wk after infection, whereas patients with severe disease continued to have elevated postoperative MACE risk even after waiting for 8 wk. Being fully vaccinated decreased the risk for postoperative MACE in both patients with no history of COVID-19 and in those with breakthrough COVID-19 infection. Together, our results suggest that a thorough assessment of the severity, vaccination status, and timing of SARS-CoV-2 infection must be a mandatory part of perioperative stratification.NEW & NOTEWORTHY With an increasing proportion of patients undergoing surgery with a prior history of COVID-19, it is crucial to understand the impact of SARS-CoV-2 infection on postoperative cardiovascular/cerebrovascular risk. Our work assesses a large, national, multi-institutional cohort of patients to highlight that COVID-19 infection increases risk for postoperative major adverse cardiovascular and cerebrovascular events (MACE). MACE risk is influenced by an interplay between disease severity and time between infection and surgery, and full vaccination reduces the risk for 30-day postoperative MACE. These results highlight the importance of stratifying time-to-surgery guidelines based on disease severity.

Assessing the Complex General Surgical Oncology Pipeline: Trends in Race and Ethnicity Among US Medical Students, General Surgery Residents, and Complex General Surgical Oncology Trainees.

BACKGROUND: Cancer incidence is expected to increase in coming decades, disproportionately so among minoritized communities. Racially and ethnically concordant care is essential to addressing disparities in cancer outcomes within at-risk groups. Here, we assess trends in racial and ethnic representation of medical students (MS), general surgery (GS) residents, and complex general surgical oncology (CGSO) fellows. METHODS: This is a retrospective review of data from the American Association of Medical Colleges and the Accreditation Council of Medical Education (ACGME) from 2015 to 2020. Self-reported race and ethnicity was obtained for MS, GS, and CGSO trainees. Race and ethnicity proportions were compared with respective representation in the 2020 US Census. Mann-Kendall, Wilcoxon rank sum, and linear regression were used to assess trends, as appropriate. RESULTS: A total of 316,448 MS applicants, 128,729 MS matriculants, 27,574 GS applicants, 46,927 active GS residents, 710 CGSO applicants, and 659 active CGSO fellows were included. With every progressive stage in training, there was a smaller proportion of URM active trainees than applicants. Further, URM, Hispanic/Latino, and Black/African American trainees were significantly underrepresented compared with 2020 Census data. While the proportion of White CGSO fellows increased over time (54.5-69.2%, p = 0.009), the proportion of Black/African American and Hispanic/Latino (URM) CGSO fellows did not significantly change over the study period, though URM representation was lower in 2020 as compared with 2015. DISCUSSION: From 2015 to 2020, minority representation decreased at every advancing stage in surgical oncology training. Efforts to address barriers for URM applicants to CGSO fellowships are needed.

Clinical Impact and Accuracy of Shave Biopsy for Initial Diagnosis of Cutaneous Melanoma.

INTRODUCTION: Effective treatment of malignant melanomas is dependent upon accurate histopathological staging of preoperative biopsy specimens. While narrow excision is the gold standard for melanoma diagnosis, superficial shave biopsies have become the preferred method by dermatologists but may transect the lesion and result in inaccurate Breslow thickness assessment. This is a retrospective cohort study evaluating an initial method of biopsy for diagnosis of cutaneous melanoma and indication for reoperation based on inaccurate initial T-staging. METHODS: We retrospectively analyzed consecutive patients referred to the Medical College of Wisconsin, a tertiary cancer center, with a diagnosis of primary cutaneous melanoma. Adult patients seen between 2015 and 2018 were included. Fisher's exact test was used to assess the association between method of initial biopsy and need for unplanned reoperation. RESULTS: Three hundred twenty three patients with cutaneous melanoma from the head and neck (H&N, n = 101, 31%), trunk (n = 90, 15%), upper extremity (n = 84, 26%), and lower extremity (n = 48, 28%) were analyzed. Median Breslow thickness was 0.54 mm (interquartile range = 0.65). Shave biopsy was the method of initial biopsy in 244 (76%), excision in 23 (7%), and punch biopsy in 56 (17%). Thirty nine (33%) shave biopsies had a positive deep margin, as did seven (23%) punch biopsies and 0 excisional biopsies. Residual melanoma at definitive excision was found in 131 (42.5%) of all surgical specimens: 95 (40.6%) shave biopsy patients, 32 (60.4%) punch biopsy patients, and four (19.0%) excision biopsy patients. Recommendations for excision margin or sentinel lymph node biopsy changed in 15 (6%) shave biopsy patients and five (9%) punch biopsy patients. CONCLUSIONS: Shave biopsy is the most frequent method of diagnosis of cutaneous melanoma in the modern era. While shave and punch biopsies may underestimate true T-stage, there was no difference in need for reoperation due to T-upstaging based on initial biopsy type, supporting current diagnostic practices. Partial biopsies can thus be used to guide appropriate treatment and definitive wide local excision when adjusting for understaging.

Opioid system is necessary but not sufficient for antidepressive actions of ketamine in rodents.

Slow response to the standard treatment for depression increases suffering and risk of suicide. Ketamine, an N-methyl-d-aspartate (NMDA) receptor antagonist, can rapidly alleviate depressive symptoms and reduce suicidality, possibly by decreasing hyperactivity in the lateral habenula (LHb) brain nucleus. Here we find that in a rat model of human depression, opioid antagonists abolish the ability of ketamine to reduce the depression-like behavioral and LHb hyperactive cellular phenotypes. However, activation of opiate receptors alone is not sufficient to produce ketamine-like effects, nor does ketamine mimic the hedonic effects of an opiate, indicating that the opioid system does not mediate the actions of ketamine but rather is permissive. Thus, ketamine does not act as an opiate but its effects require both NMDA and opiate receptor signaling, suggesting that interactions between these two neurotransmitter systems are necessary to achieve an antidepressant effect.

Benign mature teratoma of the cecum: a case report.

A teratoma is a typically benign tumor derived from more than one embryonic cell line, and it is characterized by presence of tissue foreign to the tumor location site. With the unlikely primary location in the gastrointestinal tract and no history of malignancy, we present a rare case of a primary mature cystic teratoma of the cecum. The patient is a 66-year-old male with imaging demonstrating an extraluminal, seemingly fat-containing mass abutting the cecum. The patient underwent resection, and final pathology revealed a mature cystic teratoma. Primary mature teratoma of the cecum is exceptionally rare; thus, diagnosis can be challenging. As he had no primary testicular or retroperitoneal mass, this cystic lesion likely represents a developmental abnormality and not a true neoplasm. The radiographic features, presentation, differential diagnoses, and treatment recommendations are discussed.

Automated machine learning (AutoML) can predict 90-day mortality after gastrectomy for cancer.

Early postoperative mortality risk prediction is crucial for clinical management of gastric cancer. This study aims to predict 90-day mortality in gastric cancer patients undergoing gastrectomy using automated machine learning (AutoML), optimize models for preoperative prediction, and identify factors influential in prediction. National Cancer Database was used to identify stage I-III gastric cancer patients undergoing gastrectomy between 2004 and 2016. 26 features were used to train predictive models using H2O.ai AutoML. Performance on validation cohort was measured. In 39,108 patients, 90-day mortality rate was 8.8%. The highest performing model was an ensemble (AUC = 0.77); older age, nodal ratio, and length of inpatient stay (LOS) following surgery were most influential for prediction. Removing the latter two parameters decreased model performance (AUC 0.71). For optimizing models for preoperative use, models were developed to first predict node ratio or LOS, and these predicted values were inputted for 90-day mortality prediction (AUC of 0.73-0.74). AutoML performed well in predicting 90-day mortality in a larger cohort of gastric cancer patients that underwent gastrectomy. These models can be implemented preoperatively to inform prognostication and patient selection for surgery. Our study supports broader evaluation and application of AutoML to guide surgical oncologic care.