Search

Objectives: The presence of nodal metastases in patients with papillary thyroid carcinoma (PTC) has both staging and treatment implications. However, lymph nodes are often not removed during thyroidectomy. Prior work has demonstrated the capability of artificial intelligence (AI) to predict the presence of nodal metastases in PTC based on the primary tumor histopathology alone. This study aimed to replicate these results with multi-institutional data. Methods: Cases of conventional PTC were identified from the records of 2 large academic institutions. Only patients with complete pathology data, including at least 3 sampled lymph nodes, were included in the study. Tumors were designated "positive" if they had at least 5 positive lymph node metastases. First, algorithms were trained separately on each institution's data and tested independently on the other institution's data. Then, the data sets were combined and new algorithms were developed and tested. The primary tumors were randomized into 2 groups, one to train the algorithm and another to test it. A low level of supervision was used to train the algorithm. Board-certified pathologists annotated the slides. HALO-AI convolutional neural network and image software was used to perform training and testing. Receiver operator characteristic curves and the Youden J statistic were used for primary analysis. Results: There were 420 cases used in analyses, 45% of which were negative. The best performing single institution algorithm had an area under the curve (AUC) of 0.64 with a sensitivity and specificity of 65% and 61% respectively, when tested on the other institution's data. The best performing combined institution algorithm had an AUC of 0.84 with a sensitivity and specificity of 68% and 91% respectively. Conclusion: A convolutional neural network can produce an accurate and robust algorithm that is capable of predicting nodal metastases from primary PTC histopathology alone even in the setting of multi-institutional data. [ABSTRACT FROM AUTHOR]

Background: Investigate the ability of frailty status to predict post-surgical outcomes in patients with cutaneous malignancies of the scalp and neck undergoing flap reconstruction. Methods: National Surgical Quality Improvement Program database was used to isolate patients with cutaneous malignancies of the scalp and neck who underwent surgical resection between 2015 to 2019. Univariate and multivariate analyses were performed to determine if frailty score correlated with negative post-operative outcomes. Receiver operating characteristic (ROC) curves allowed testing of the discriminative performance of age versus frailty. Results: This study demonstrated an independent correlation between frailty and major complications as well as non-home discharge. In ROC curve analysis, frailty demonstrated superior discrimination compared to age for predicting major complications. Conclusion: Our study demonstrated an association between increasing frailty and major complications as well as the likelihood of a non-home discharge. When compared to age, frailty was also shown to be a better predictor of major complications. [ABSTRACT FROM AUTHOR]

Background: Anatomic severity grade (ASG) can be used to assess abdominal aortic aneurysm (AAA) anatomic complexity. High ASG is associated with complications following endovascular repair of AAAs and we have demonstrated that ASG correlates with resource utilization. The hypothesis of this study is that ASG is directly related to midterm mortality in patients undergoing AAA repair. Methods: Patients who underwent infrarenal AAA repairs between July 2007 and August 2014 were retrospectively reviewed and ASG scores were calculated using 3-dimensional computed tomography reconstructions. Perioperative mortalities (≤30 days) were excluded. The ASG value of 15 was chosen based on previous receiver–operator curve analysis, which showed that an ASG of 15 was predictive of postoperative complications and resource utilization. The 5-year survivors and mortalities were compared utilizing comorbidities, pharmacologic variables, and anatomic variables at or above the defined threshold. Results: A total of 402 patients (80% male and 96% Caucasian) with complete anatomic and survival data were included in the analysis. Mean ASG and age at the time of repair were 16 ± 0.15 and 73 ± 0.43 years old, respectively. The 5-year mortality was significantly associated with ASG >15 (hazard ratio [HR]: 1.504, confidence interval [CI]: 1.077-2.100, P <.017), hyperlipidemia (HR: 1.987, CI: 1.341-2.946, P <.001), coronary artery disease (HR: 1.432, CI: 1.037-1.978, P <.029), and chronic obstructive pulmonary disease (HR: 1.412, CI: 1.027-1.943, P <.034). Kaplan-Meier analysis demonstrated improved survival in the low score ASG ≤15 group at 1, 3, and 5 years (96% vs 93%, 81% vs 69%, and 53% vs 41%; P =.0182; Figure 1). Conclusions: Increasing aortic anatomic complexity as characterized by ASG >15 is an independent predictor of midterm mortality following elective infrarenal AAA repair. Therefore, it may be a useful tool for appropriate patient selection and risk stratification prior to elective infrarenal AAA repair. [ABSTRACT FROM AUTHOR]