Renal metastasis from esophageal adenocarcinoma: A rare recurrence.

Esophageal cancer, consisting primarily of squamous cell carcinoma and adenocarcinoma pathology, is a leading cause of morbidity and mortality worldwide with rates of metastasis at time of diagnosis up to 50%. Renal metastasis is rare, with most pathological diagnosis yielding squamous cell carcinoma. We present the unique case of a 78-year-old man with biopsy proven adenocarcinoma metastasis to the kidney on routine surveillance following initial esophagectomy, chemoradiation and adjuvant immunotherapy. Imaging features of the solitary renal metastasis highly mimicked a primary renal cell carcinoma. Additional unique features included renal pelvis invasion and disease recurrence despite adjuvant immunotherapy. This case underscores the role of routine surveillance in this patient population, varied radiologic appearance, and importance for pathologic diagnosis.

Right-to-left ventricle ratio determined by machine learning algorithms on CT pulmonary angiography images predicts prolonged ICU length of stay in operated chronic thromboembolic pulmonary hypertension.

OBJECTIVE: Right-to-left ventricle diameter ratio (dRV/dLV) on CT pulmonary angiography (CTPA) is a predictor of outcomes in non-operated chronic thromboembolic pulmonary hypertension (CTEPH) patients. The purpose of this study is to evaluate the performance of a novel machine learning (ML) algorithm for dRV/dLV measurement in operated CTEPH patients and its association with post-operative outcomes. METHODS: This retrospective study reviewed consecutive CTEPH patients who underwent pulmonary endarterectomy between 2013 and 2017. ML calculated dRV/dLV on pre-operative CTPA and compared with manual measures. Associations of dRV/dLV with patient characteristics and post-operative outcomes were evaluated including intensive care (ICU) and hospital length of stay (LOS) using multivariable linear regression analysis. Prolonged LOS was defined as greater than median. RESULTS: ML segmented the ventricles in 99/125 (79%) patients. The most common cause of failure was misidentification of the moderator band as the interventricular septum (7.9%). Mean dRV/dLV by ML was 1.4 ± 0.4 and strongly correlated with manual measures (r = 0.9-0.96 p < 0.0001). dRV/dLV was moderately correlated with measures of pulmonary hypertension on right heart catheterization and RV dilatation on echocardiogram (r = 0.5-0.6, p < 0.0001). dRV/dLV ≥ 1.2 was associated with proximal Jamieson type disease (p = 0.032), longer cardiopulmonary bypass (p = 0.037), aortic cross-clamp (p = 0.022) and circulatory arrest (p < 0.001) at surgery and dRV/dLV ≥ 1.6 with post-operative ECMO (p = 0.006). dRV/dLV was independently associated with prolonged ICU LOS (OR = 3.79, 95% CI 1.1-13.06, p = 0.035). CONCLUSION: dRV/dLV was associated with CTEPH severity and independently associated with prolonged ICU LOS. This CT parameter may therefore assist in perioperative planning. Further refinement of the ML algorithm or CTPA technique is required to avoid errors in ventricular segmentation. ADVANCES IN KNOWLEDGE: Automated right-to-left ventricle ratio measurement by machine learning is feasible and is independently associated with outcome after pulmonary endarterectomy.

Quantifying uncertainty in machine learning classifiers for medical imaging.

PURPOSE: Machine learning (ML) models in medical imaging (MI) can be of great value in computer aided diagnostic systems, but little attention is given to the confidence (alternatively, uncertainty) of such models, which may have significant clinical implications. This paper applied, validated, and explored a technique for assessing uncertainty in convolutional neural networks (CNNs) in the context of MI. MATERIALS AND METHODS: We used two publicly accessible imaging datasets: a chest x-ray dataset (pneumonia vs. control) and a skin cancer imaging dataset (malignant vs. benign) to explore the proposed measure of uncertainty based on experiments with different class imbalance-sample sizes, and experiments with images close to the classification boundary. We also further verified our hypothesis by examining the relationship with other performance metrics and cross-checking CNN predictions and confidence scores with an expert radiologist (available in the Supplementary Information). Additionally, bounds were derived on the uncertainty metric, and recommendations for interpretability were made. RESULTS: With respect to training set class imbalance for the pneumonia MI dataset, the uncertainty metric was minimized when both classes were nearly equal in size (regardless of training set size) and was approximately 17% smaller than the maximum uncertainty resulting from greater imbalance. We found that less-obvious test images (those closer to the classification boundary) produced higher classification uncertainty, about 10-15 times greater than images further from the boundary. Relevant MI performance metrics like accuracy, sensitivity, and sensibility showed seemingly negative linear correlations, though none were statistically significant (p [Formula: see text] 0.05). The expert radiologist and CNN expressed agreement on a small sample of test images, though this finding is only preliminary. CONCLUSIONS: This paper demonstrated the importance of uncertainty reporting alongside predictions in medical imaging. Results demonstrate considerable potential from automatically assessing classifier reliability on each prediction with the proposed uncertainty metric.

Clinical Implications of Body Composition and Exercise Capacity Following Pulmonary Endarterectomy.

BACKGROUND: Pulmonary endarterectomy (PEA) is a curative procedure for patients with chronic thromboembolic pulmonary hypertension. Body composition and exercise capacity have been associated with adverse outcomes in patients undergoing cardiothoracic operations, but their significance with PEA is unclear. We evaluated the association of body composition and 6-minute walk distance (6MWD) with disease severity, hospital length of stay, discharge disposition, and postoperative functional recovery. METHODS: This was a retrospective, single-center cohort study of patients who underwent PEA (January 2014-December 2017). Body composition (skeletal muscle mass and adiposity cross-sectional area) was quantified using thoracic computed tomography with sliceOmatic (TomoVision, Magog, QC, Canada) software. Body mass index was calculated. Association of body composition measures and 6MWD with clinical outcomes was evaluated using multivariable regression models. RESULTS: The study included 127 patients (42% men), aged 58 ± 14 years; body mass index was 31 ± 7 kg/m2 and 6MWD was 361 ± 165 m). Muscle and 6MWD were associated with disease severity measures. Of those surviving hospitalization (n = 125), a greater 6MWD was associated with a shorter hospital stay (1.9 median days per 100 m; p < .001) and higher likelihood of being discharged directly home from hospital (odds ratio, 2.1 per 100 m; P = .004), independent of age, sex, and body mass index. Those with a lower preoperative 6MWD (per 100 m) had a greater increase in their postoperative 6MWD (52 m; P < .0001), independent of age, sex, and body mass index. Body composition measures were not associated with hospital outcomes or exercise capacity in the first year postoperatively. CONCLUSIONS: Exercise capacity was a more prognostic marker of PEA outcomes compared with body composition. Future research is needed to explore pre-PEA rehabilitation strategies.

Effect of Perioperative Active Body Surface Warming Systems on Analgesic and Clinical Outcomes: A Systematic Review and Meta-analysis of Randomized Controlled Trials.

BACKGROUND: Inadvertent perioperative hypothermia is a common complication of surgery, and active body surface warming (ABSW) systems are used to prevent adverse clinical outcomes. Prior data on certain outcomes are equivocal (ie, blood loss) or limited (ie, pain and opioid consumption). The objective of this study was to provide an updated review on the effect of ABSW on clinical outcomes and temperature maintenance. METHODS: We conducted a systematic review of randomized controlled trials evaluating ABSW systems compared to nonactive warming controls in noncardiac surgeries. Outcomes studied included postoperative pain scores and opioid consumption (primary outcomes) and other perioperative clinical variables such as temperature changes, blood loss, and wound infection (secondary outcomes). We searched Ovid MEDLINE daily, Ovid MEDLINE, EMBASE, CINHAL, Cochrane CENTRAL, and Web of Science from inception to June 2019. Quality of evidence (QoE) was rated according to the Grading of Recommendations, Assessment, Development and Evaluations (GRADE) approach. Subgroup analysis sought to determine the effect of preoperative + intraoperative warming versus intraoperative warming alone. Metaregression evaluated the effect of year of publication, use of neuromuscular blockers, anesthesia, and surgery type on outcomes. RESULTS: Fifty-four articles (3976 patients) were included. Pooled results demonstrated that ABSW maintained normothermia compared to controls, during surgery (30 minutes postinduction [mean difference {MD}: 0.3°C, 95% confidence interval {CI}, 0.2-0.4, moderate QoE]), end of surgery (MD: 1.1°C, 95% CI, 0.9-1.3, high QoE), and up to 4 hours postoperatively (MD: 0.3°C, 95% CI, 0.2-0.5, high QoE). ABSW was not associated with difference in pain scores (<24 hours postoperatively, moderate to low QoE) or perioperative opioid consumption (very low QoE). ABSW increased patient satisfaction (MD: 2.2 points, 95% CI, 0.9-3.6, moderate QoE), reduced blood transfusions (odds ratio [OR] = 0.6, 95% CI, 0.4-1.0, moderate QoE), shivering (OR = 0.2, 95% CI, 0.1-0.4, high QoE), and wound infections (OR = 0.3, 95% CI, 0.2-0.7, high QoE). No significant differences were found for fluid administration (low QoE), blood loss (very low QoE), major adverse cardiovascular events (very low QoE), or mortality (very low QoE). Subgroup analysis and metaregression suggested increased temperature benefit with pre + intraoperative warming, use of neuromuscular blockers, and recent publication year. ABSW seemed to confer less temperature benefit in cesarean deliveries and neurosurgical/spinal cases compared to abdominal surgeries. CONCLUSIONS: ABSW is effective in maintaining physiological normothermia, decreasing wound infections, shivering, blood transfusions, and increasing patient satisfaction but does not appear to affect postoperative pain and opioid use.

Assessing and mitigating the effects of class imbalance in machine learning with application to X-ray imaging.

PURPOSE: Machine learning (ML) algorithms are well known to exhibit variations in prediction accuracy when provided with imbalanced training sets typically seen in medical imaging (MI) due to the imbalanced ratio of pathological and normal cases. This paper presents a thorough investigation of the effects of class imbalance and methods for mitigating class imbalance in ML algorithms applied to MI. METHODS: We first selected five classes from the Image Retrieval in Medical Applications (IRMA) dataset, performed multiclass classification using the random forest model (RFM), and then performed binary classification using convolutional neural network (CNN) on a chest X-ray dataset. An imbalanced class was created in the training set by varying the number of images in that class. Methods tested to mitigate class imbalance included oversampling, undersampling, and changing class weights of the RFM. Model performance was assessed by overall classification accuracy, overall F1 score, and specificity, recall, and precision of the imbalanced class. RESULTS: A close-to-balanced training set resulted in the best model performance, and a large imbalance with overrepresentation was more detrimental to model performance than underrepresentation. Oversampling and undersampling methods were both effective in mitigating class imbalance, and efficacy of oversampling techniques was class specific. CONCLUSION: This study systematically demonstrates the effect of class imbalance on two public X-ray datasets on RFM and CNN, making these findings widely applicable as a reference. Furthermore, the methods employed here can guide researchers in assessing and addressing the effects of class imbalance, while considering the data-specific characteristics to optimize imbalance mitigating methods.

Effect of seizure timing on long-term survival in patients with brain tumor.

OBJECTIVE: Seizures often occur in patients with primary brain tumor (BT). The aim of this study was to determine if there is an association between the time of occurrence of seizures during the course of BT and survival of these patients. METHODS: This retrospective cohort study at Henry Ford Hospital, an urban tertiary referral center, included all patients who were diagnosed with primary BTs at Henry Ford Health System between January 2006 and December 2014. Timing of seizure occurrence, if occurred at presentation or after the tumor diagnosis during follow-up period, in different grades of BTs, and survival of these patients were analyzed. RESULTS: Of the 901 identified patients, 662 (53% male; mean age: 56 years) were included in final analysis, and seizures occurred in 283 patients (43%). Patients with World Health Organization (WHO) grade III BT with seizures as a presenting symptom only had better survival (adjusted hazard ratio (HR): 0.27; 95% confidence interval (CI), 0.11-0.67; P = 0.004). Seizures that occurred after tumor diagnosis only (adjusted HR: 2.11; 95% CI, 1.59-2.81; P < 0.001) in patients with WHO grade II tumors (adjusted HR: 3.41; 95% CI, 1.05-11.1; P = 0.041) and WHO grade IV tumors (adjusted HR: 2.14; 95% CI, 1.58-2.90; P < 0.001) had higher mortality. Seizures that occurred at presentation and after diagnosis also had higher mortality (adjusted HR: 1.34; 95% CI, 1.00-1.80; P = 0.049), in patients with meningioma (adjusted HR: 6.19; 95% CI, 1.30-29.4; P = 0.021) and grade III tumors (adjusted HR: 6.19; 95% CI, 2.56-15.0; P < 0.001). CONCLUSION: Seizures occurred in almost half of the patients with BTs. The association between seizures in patients with BT and their survival depends on the time of occurrence of seizures, if occurring at presentation or after tumor diagnosis, and the type of tumor. Better survival was noted in patients with WHO grade III BTs who had seizures at presentation at the time of diagnosis, while higher mortality was noted in WHO grade II tumors who had seizure at presentation and after tumor diagnosis, and in grade IV tumors after tumor diagnosis.