Environmentally sustainable design guide for haemodialysis facilities: An Australian and New Zealand society of Nephrology initiative.

Haemodialysis facilities have a large environmental impact due to high energy, water and consumable usage by haemodialysis equipment. As climate change and natural resource scarcity escalate, all the while the number of people requiring dialysis increases, there is an urgent need for dialysis facilities that meet care needs while minimising environmental impact. To address this, the Australian and New Zealand Society of Nephrology engaged an environmental sustainability consulting practise to develop a best practise guide for the environmentally sustainable design and operation of haemodialysis facilities. Four opportunity areas were considered, namely energy, water, waste and resource recovery, and additional sustainability. A total of 28 environmental improvement initiatives were identified. The majority (n = 23) were general measures that could be applied across all healthcare settings, while five were specific to haemodialysis facilities. Recommendations were made regarding specific measures that should be undertaken and/or standards that must be met to achieve the intent of each initiative. These were stratified to enable their application to both existing dialysis facilities and new builds. The lifecycle stage of a haemodialysis facility to which each initiative applied was highlighted, as was its potential impact. This guide provides a tailored and comprehensive resource for the kidney care community to enable the integration of best practise sustainability considerations into both existing and new facilities. If broadly implemented, it has the potential to markedly improve the environmental impact of haemodialysis provision.

Functional Connectivity Changes on Resting-State fMRI after Mild Traumatic Brain Injury: A Systematic Review.

BACKGROUND: Mild traumatic brain injury is theorized to cause widespread functional changes to the brain. Resting-state fMRI may be able to measure functional connectivity changes after traumatic brain injury, but resting-state fMRI studies are heterogeneous, using numerous techniques to study ROIs across various resting-state networks. PURPOSE: We systematically reviewed the literature to ascertain whether adult patients who have experienced mild traumatic brain injury show consistent functional connectivity changes on resting-state -fMRI, compared with healthy patients. DATA SOURCES: We used 5 databases (PubMed, EMBASE, Cochrane Central, Scopus, Web of Science). STUDY SELECTION: Five databases (PubMed, EMBASE, Cochrane Central, Scopus, and Web of Science) were searched for research published since 2010. Search strategies used keywords of "functional MR imaging" and "mild traumatic brain injury" as well as related terms. All results were screened at the abstract and title levels by 4 reviewers according to predefined inclusion and exclusion criteria. For full-text inclusion, each study was evaluated independently by 2 reviewers, with discordant screening settled by consensus. DATA ANALYSIS: Data regarding article characteristics, cohort demographics, fMRI scan parameters, data analysis processing software, atlas used, data characteristics, and statistical analysis information were extracted. DATA SYNTHESIS: Across 66 studies, 80 areas were analyzed 239 times for at least 1 time point, most commonly using independent component analysis. The most analyzed areas and networks were the whole brain, the default mode network, and the salience network. Reported functional connectivity changes varied, though there may be a slight trend toward decreased whole-brain functional connectivity within 1 month of traumatic brain injury and there may be differences based on the time since injury. LIMITATIONS: Studies of military, sports-related traumatic brain injury, and pediatric patients were excluded. Due to the high number of relevant studies and data heterogeneity, we could not be as granular in the analysis as we would have liked. CONCLUSIONS: Reported functional connectivity changes varied, even within the same region and network, at least partially reflecting differences in technical parameters, preprocessing software, and analysis methods as well as probable differences in individual injury. There is a need for novel rs-fMRI techniques that better capture subject-specific functional connectivity changes.

Terminal Ileum Lipoma Causing Ileocolic Intussusception: A Case Report and Literature Review.

Adult intussusception is much rarer than pediatric intussusception and usually occurs secondary to a pathological lead point, most frequently neoplasm. Terminal ileum lipomas are an infrequent cause of adult ileocolic intussusception but can be seen together with the intussusception on initial imaging evaluation, which can guide appropriate diagnosis and management. We describe a case of a 42-year-old man presenting with 12 hours of severe right lower quadrant pain. CT of the abdomen and pelvis demonstrated an ileocolic intussusception with fat-density lesions within the intussusception as well as in the distal ileum. The patient went to the operating room for laparoscopic ileocolic resection, during which ileo-ileal and ileocolic intussusceptions were identified in the terminal ileum and multiple fatty masses were palpated in the terminal ileum and cecum. Following ileocecectomy, surgical pathology confirmed terminal ileum with intussusception associated with multiple submucosal lipomas. We also review the literature for cases of ileocolic intussusception caused by terminal ileum lipomas. Patients presented with both acute and chronic symptoms, and while CT was the most common modality used for diagnosis, ultrasound and colonoscopy were also able to identify the intussusception. Although the intussusception was initially reduced in two patients, all patients ultimately underwent surgical resection.

Assessment of Contact Allergens in "Hypoallergenic" Athletic Shoes by Mass Spectrometry.

Background: Identification of athletic shoes for patients with contact allergy is difficult. Company reports of allergen content are often incorrect. Objectives: To determine whether chemical analysis of 4 athletic shoes, previously reported to be free of the most common contact allergens, contain quantifiable allergen levels. Methods: Samples from the uppers and insoles of 4 shoes believed to be free of common allergens were assessed by mass spectrometry. A total of 4 rubber accelerators and 2 adhesives were directly quantified and additional 7 rubber accelerators were assessed using semiquantitative measures. Results: Aside from carbamates (assayed as 59 ppm zinc in insoles) in SeaVee's Sixty-Six sneakers, para-tertiarybutylphenol formaldehyde resin (PTBFR) (assayed as 7.6 ppm paratertiary butylphenol or 4-tertiary butylphenol [4TBP] in uppers) in Allbirds Tree Runners and rosin (assayed as 628 ppm sodium abietate in uppers) and carbamates (24 ppm zinc in uppers) in Saucony Jazz sneakers, these shoes had low levels of all allergens assayed in this study. Tom's Carlo sneakers contained rosin (127 ppm sodium abietate in insoles), PTBFR (6.5 ppm 4TBP in uppers), and carbamates (112 ppm sodium abietate in insoles) but had low levels of all other assayed allergens. Conclusions: Although identifying allergen-free shoes is challenging, the results of this analysis will help patch testing physicians recommend athletic shoes to patients with specific allergies.

Large language models encode clinical knowledge.

Large language models (LLMs) have demonstrated impressive capabilities, but the bar for clinical applications is high. Attempts to assess the clinical knowledge of models typically rely on automated evaluations based on limited benchmarks. Here, to address these limitations, we present MultiMedQA, a benchmark combining six existing medical question answering datasets spanning professional medicine, research and consumer queries and a new dataset of medical questions searched online, HealthSearchQA. We propose a human evaluation framework for model answers along multiple axes including factuality, comprehension, reasoning, possible harm and bias. In addition, we evaluate Pathways Language Model1 (PaLM, a 540-billion parameter LLM) and its instruction-tuned variant, Flan-PaLM2 on MultiMedQA. Using a combination of prompting strategies, Flan-PaLM achieves state-of-the-art accuracy on every MultiMedQA multiple-choice dataset (MedQA3, MedMCQA4, PubMedQA5 and Measuring Massive Multitask Language Understanding (MMLU) clinical topics6), including 67.6% accuracy on MedQA (US Medical Licensing Exam-style questions), surpassing the prior state of the art by more than 17%. However, human evaluation reveals key gaps. To resolve this, we introduce instruction prompt tuning, a parameter-efficient approach for aligning LLMs to new domains using a few exemplars. The resulting model, Med-PaLM, performs encouragingly, but remains inferior to clinicians. We show that comprehension, knowledge recall and reasoning improve with model scale and instruction prompt tuning, suggesting the potential utility of LLMs in medicine. Our human evaluations reveal limitations of today's models, reinforcing the importance of both evaluation frameworks and method development in creating safe, helpful LLMs for clinical applications.

Analysis of Factors Potentially Influencing Diagnostic Yield Among Computed Tomography-Guided Biopsies for Bone Lesions.

OBJECTIVE: This study aimed to investigate patient-specific, lesion-related, and technical factors that potentially influence diagnostic yield of computed tomography (CT)-guided biopsies of bone lesions. METHODS: Computed tomography-guided bone lesion biopsies performed over a 2-year period were retrospectively reviewed, including image review and electronic medical records for pathology reports and clinical follow-up. Lesions were tabulated by prebiopsy CT and magnetic resonance imaging features. Patients with nondiagnostic biopsies were assessed for presumptive clinical diagnosis and management. RESULTS: Nondiagnostic pathology results were obtained in 31 of 156 cases (19.87%), among which diagnoses were confirmed by other tissue sampling in 9; clinical follow-up of up to 2 years yielded no diagnosis in 10 and presumptive diagnoses in 12. The nondiagnostic biopsy rate of long bone lesions was higher than that of other bone lesions (odds ratio, 3.46; 95% confidence interval, 1.32-9.09). There were no significant differences in patient American Society of Anesthesiologists class, mean body mass index, sedation method, number of cores, or needle gauge between diagnostic and nondiagnostic biopsy cohorts. Diagnostic yield was not significantly different between occult, lytic, or sclerotic lesions. There was no difference in diagnostic yield regarding presence of cortical break, gadolinium enhancement, or lesion depth. Magnetic resonance imaging was obtained before biopsy in significantly more nondiagnostic cases compared with diagnostic cases ( P = 0.027). CONCLUSIONS: Computed tomography-guided biopsies had a nondiagnostic rate of 19.87%, and lesions in the long bones of the extremities were disproportionately common among this group. There was no significant association between biopsy results and several patient-specific, lesion-related, and technical factors.

Personalized multitarget biologic injection in the spine: prospective case series of multitarget platelet-rich plasma for low back pain.

Aim: To evaluate the use of a multitarget platelet-rich plasma (PRP) injection approach for the treatment of chronic low back pain (LBP). Materials & Methods: Forty-six patients with more than 12 weeks of LBP who failed conservative treatments were injected with PRP into the facet joints, intervertebral discs, epidural space and/or paravertebral muscles. Visual analog pain scale and Roland-Morris Disability Questionnaire scores were measured at baseline and predefined intervals. Results: Mean visual analog pain scale was reduced from 8.48 to 5.17 and mean Roland-Morris Disability Questionnaire from 18.0 to 10.98 at 12 weeks (p < 0.001). These statistically significant improvements were sustained over 52 weeks. No adverse effects were observed. Conclusion: Our PRP approach demonstrated clinically favorable results and may be a promising treatment for chronic LBP.

Lay abstract Back pain can be caused by a variety of conditions. Most long-term (chronic) low back pain cases involve one or more parts of the spine causing the pain. This study describes 46 people who received injections of a blood-based substance called platelet-rich plasma into multiple parts of their spine to address chronic low back pain. The patients were followed up at several time points over the course of the following year. The results showed that the patients had improvement in their pain and disability. There was also a reduction in oral pain medication use. No unexpected medical problems were seen with this treatment. This study shows promising results for the treatment of chronic back pain.

Evaluation of an Artificial Intelligence-Augmented Digital System for Histologic Classification of Colorectal Polyps.

Importance: Colorectal polyps are common, and their histopathologic classification is used in the planning of follow-up surveillance. Substantial variation has been observed in pathologists' classification of colorectal polyps, and improved assessment by pathologists may be associated with reduced subsequent underuse and overuse of colonoscopy. Objective: To compare standard microscopic assessment with an artificial intelligence (AI)-augmented digital system that annotates regions of interest within digitized polyp tissue and predicts polyp type using a deep learning model to assist pathologists in colorectal polyp classification. Design, Setting, and Participants: In this diagnostic study conducted at a tertiary academic medical center and a community hospital in New Hampshire, 100 slides with colorectal polyp samples were read by 15 pathologists using a microscope and an AI-augmented digital system, with a washout period of at least 12 weeks between use of each modality. The study was conducted from February 10 to July 10, 2020. Main Outcomes and Measures: Accuracy and time of evaluation were used to compare pathologists' performance when a microscope was used with their performance when the AI-augmented digital system was used. Outcomes were compared using paired t tests and mixed-effects models. Results: In assessments of 100 slides with colorectal polyp specimens, use of the AI-augmented digital system significantly improved pathologists' classification accuracy compared with microscopic assessment from 73.9% (95% CI, 71.7%-76.2%) to 80.8% (95% CI, 78.8%-82.8%) (P < .001). The overall difference in the evaluation time per slide between the digital system (mean, 21.7 seconds; 95% CI, 20.8-22.7 seconds) and microscopic examination (mean, 13.0 seconds; 95% CI, 12.4-13.5 seconds) was -8.8 seconds (95% CI, -9.8 to -7.7 seconds), but this difference decreased as pathologists became more familiar and experienced with the digital system; the difference between the time of evaluation on the last set of 20 slides for all pathologists when using the microscope and the digital system was 4.8 seconds (95% CI, 3.0-6.5 seconds). Conclusions and Relevance: In this diagnostic study, an AI-augmented digital system significantly improved the accuracy of pathologic interpretation of colorectal polyps compared with microscopic assessment. If applied broadly to clinical practice, this tool may be associated with decreases in subsequent overuse and underuse of colonoscopy and thus with improved patient outcomes and reduced health care costs.

Identifying Safe Alternatives for Contact Allergy Patients.

Allergen avoidance is the most effective treatment of contact allergy. Patient improvement ultimately relies on identification of safe alternative products, which can be used by the patient. Safe personal care product options typically can be found using ingredient database programs. Avoidance of allergens in other products (eg, shoes, clothing, and dental care) often is challenging. This article discusses specific safe alternatives for the 80 allergens on the 2017 American Contact Dermatitis Society core allergen series.

Safety and Effectiveness of Yttrium-90 Radioembolization around the Time of Immune Checkpoint Inhibitors for Unresectable Hepatic Metastases.

PURPOSE: To assess the safety and effectiveness of yttrium-90 radioembolization and checkpoint inhibitor immunotherapy within a short interval for the treatment of unresectable hepatic metastases. MATERIALS AND METHODS: This single-institution retrospective study included 22 patients (12 men; median age, 65 y ± 11) with unresectable hepatic metastases and preserved functional status (Eastern Cooperative Oncology Group performance status 0/1) who received immunotherapy and radioembolization within a 15-month period (median, 63.5 d; interquartile range, 19.7-178.2 d) from February 2013 to March 2018. Primary malignancies were uveal melanoma (12 of 22; 54.5%), soft tissue sarcoma (3; 13.6%), cutaneous melanoma (3; 14%), and others (4; 18.2%). Studies were reviewed to March 2019 to assess Common Terminology Criteria for Adverse Events grade 3/4 toxicities, disease progression, and death. RESULTS: There were no grade 4 toxicities within 6 mo of radioembolization. Grade 3 hepatobiliary toxicities occurred in 3 patients (13.6%) within 6 months, 2 from rapid disease progression and 1 from a biliary stricture. Two patients (9.1%) experienced clinical toxicities, including grade 4 colitis at 6 months and hepatic abscess at 3 months. Median overall survival (OS) from first radioembolization was 20 mo (95% confidence interval [CI], 12.5-27.5 mo), and median OS from first immunotherapy was 23 months (95% CI, 15.9-30.1 mo). Within the uveal melanoma subgroup, the median OS from first radioembolization was 17.0 months (95% CI, 14.2-19.8 mo). Median time to progression was 7.8 months (95% CI, 3.3-12.2 mo), and median progression-free survival was 7.8 mo (95% CI, 3.1-12.4 mo). CONCLUSIONS: Checkpoint immunotherapy around the time of radioembolization is safe, with a low incidence of toxicity independent of primary malignancy.

Predicting colorectal polyp recurrence using time-to-event analysis of medical records.

Identifying patient characteristics that influence the rate of colorectal polyp recurrence can provide important insights into which patients are at higher risk for recurrence. We used natural language processing to extract polyp morphological characteristics from 953 polyp-presenting patients' electronic medical records. We used subsequent colonoscopy reports to examine how the time to polyp recurrence (731 patients experienced recurrence) is influenced by these characteristics as well as anthropometric features using Kaplan-Meier curves, Cox proportional hazards modeling, and random survival forest models. We found that the rate of recurrence differed significantly by polyp size, number, and location and patient smoking status. Additionally, right-sided colon polyps increased recurrence risk by 30% compared to left-sided polyps. History of tobacco use increased polyp recurrence risk by 20% compared to never-users. A random survival forest model showed an AUC of 0.65 and identified several other predictive variables, which can inform development of personalized polyp surveillance plans.

Pathophysiological Mechanisms by which Heat Stress Potentially Induces Kidney Inflammation and Chronic Kidney Disease in Sugarcane Workers.

BACKGROUND: Chronic kidney disease of non-traditional origin (CKDnt) is common among Mesoamerican sugarcane workers. Recurrent heat stress and dehydration is a leading hypothesis. Evidence indicate a key role of inflammation. METHODS: Starting in sports and heat pathophysiology literature, we develop a theoretical framework of how strenuous work in heat could induce kidney inflammation. We describe the release of pro-inflammatory substances from a leaky gut and/or injured muscle, alone or in combination with tubular fructose and uric acid, aggravation by reduced renal blood flow and increased tubular metabolic demands. Then, we analyze longitudinal data from >800 sugarcane cutters followed across harvest and review the CKDnt literature to assess empirical support of the theoretical framework. RESULTS: Inflammation (CRP elevation and fever) and hyperuricemia was tightly linked to kidney injury. Rehydrating with sugary liquids and NSAID intake increased the risk of kidney injury, whereas electrolyte solution consumption was protective. Hypokalemia and hypomagnesemia were associated with kidney injury. DISCUSSION: Heat stress, muscle injury, reduced renal blood flow and fructose metabolism may induce kidney inflammation, the successful resolution of which may be impaired by daily repeating pro-inflammatory triggers. We outline further descriptive, experimental and intervention studies addressing the factors identified in this study.

Evaluation of a Deep Neural Network for Automated Classification of Colorectal Polyps on Histopathologic Slides.

Importance: Histologic classification of colorectal polyps plays a critical role in screening for colorectal cancer and care of affected patients. An accurate and automated algorithm for the classification of colorectal polyps on digitized histopathologic slides could benefit practitioners and patients. Objective: To evaluate the performance and generalizability of a deep neural network for colorectal polyp classification on histopathologic slide images using a multi-institutional data set. Design, Setting, and Participants: This prognostic study used histopathologic slides collected from January 1, 2016, to June 31, 2016, from Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire, with 326 slides used for training, 157 slides for an internal data set, and 25 for a validation set. For the external data set, 238 slides for 179 distinct patients were obtained from 24 institutions across 13 US states. Data analysis was performed from April 9 to November 23, 2019. Main Outcomes and Measures: Accuracy, sensitivity, and specificity of the model to classify 4 major colorectal polyp types: tubular adenoma, tubulovillous or villous adenoma, hyperplastic polyp, and sessile serrated adenoma. Performance was compared with that of local pathologists' at the point of care identified from corresponding pathology laboratories. Results: For the internal evaluation on the 157 slides with ground truth labels from 5 pathologists, the deep neural network had a mean accuracy of 93.5% (95% CI, 89.6%-97.4%) compared with local pathologists' accuracy of 91.4% (95% CI, 87.0%-95.8%). On the external test set of 238 slides with ground truth labels from 5 pathologists, the deep neural network achieved an accuracy of 87.0% (95% CI, 82.7%-91.3%), which was comparable with local pathologists' accuracy of 86.6% (95% CI, 82.3%-90.9%). Conclusions and Relevance: The findings suggest that this model may assist pathologists by improving the diagnostic efficiency, reproducibility, and accuracy of colorectal cancer screenings.

Safety of Combined Yttrium-90 Radioembolization and Immune Checkpoint Inhibitor Immunotherapy for Hepatocellular Carcinoma.

PURPOSE: To investigate the safety of yttrium-90 radioembolization in combination with checkpoint inhibitor immunotherapy for the treatment of hepatocellular carcinoma (HCC). MATERIALS AND METHODS: This single-center retrospective study included 26 consecutive patients with HCC who received checkpoint inhibitor immunotherapy within 90 days of radioembolization from April 2015 to May 2018. Patients had preserved liver function (Child-Pugh scores A-B7) and either advanced HCC due to macrovascular invasion or limited extrahepatic disease (21 patients) or aggressive intermediate stage HCC that resulted in earlier incorporation of systemic immunotherapy (5 patients). Clinical documentation, laboratory results, and imaging results at 1- and 3-month follow-up intervals were reviewed to assess treatment-related adverse events and treatment responses. RESULTS: The median follow-up period after radioembolization was 7.8 months (95% confidence interval [CI], 5.6-11.8). There were no early (30-day) mortality or grades 3/4 hepatobiliary or immunotherapy-related toxicities. Delayed grades 3/4 hepatobiliary toxicities (1-3 months) occurred in 2 patients in the setting of HCC disease progression. One patient developed pneumonitis. The median overall survival from first immunotherapy was 17.2 months (95% CI, 10.9-23.4). The median overall survival from first radioembolization was 16.5 months (95% CI, 6.6-26.4). From first radioembolization, time to tumor progression was 5.7 months (95% CI, 4.2-7.2), and progression-free survival was 5.7 months (95% CI, 4.3-7.1). CONCLUSIONS: Radioembolization combined with checkpoint inhibitor immunotherapy in cases of HCC appears to be safe and causes limited treatment-related toxicity. Future prospective studies are needed to identify the optimal combination treatment protocols and evaluate the efficacy of combination therapy.

Attention-Based Deep Neural Networks for Detection of Cancerous and Precancerous Esophagus Tissue on Histopathological Slides.

Importance: Deep learning-based methods, such as the sliding window approach for cropped-image classification and heuristic aggregation for whole-slide inference, for analyzing histological patterns in high-resolution microscopy images have shown promising results. These approaches, however, require a laborious annotation process and are fragmented. Objective: To evaluate a novel deep learning method that uses tissue-level annotations for high-resolution histological image analysis for Barrett esophagus (BE) and esophageal adenocarcinoma detection. Design, Setting, and Participants: This diagnostic study collected deidentified high-resolution histological images (N = 379) for training a new model composed of a convolutional neural network and a grid-based attention network. Histological images of patients who underwent endoscopic esophagus and gastroesophageal junction mucosal biopsy between January 1, 2016, and December 31, 2018, at Dartmouth-Hitchcock Medical Center (Lebanon, New Hampshire) were collected. Main Outcomes and Measures: The model was evaluated on an independent testing set of 123 histological images with 4 classes: normal, BE-no-dysplasia, BE-with-dysplasia, and adenocarcinoma. Performance of this model was measured and compared with that of the current state-of-the-art sliding window approach using the following standard machine learning metrics: accuracy, recall, precision, and F1 score. Results: Of the independent testing set of 123 histological images, 30 (24.4%) were in the BE-no-dysplasia class, 14 (11.4%) in the BE-with-dysplasia class, 21 (17.1%) in the adenocarcinoma class, and 58 (47.2%) in the normal class. Classification accuracies of the proposed model were 0.85 (95% CI, 0.81-0.90) for the BE-no-dysplasia class, 0.89 (95% CI, 0.84-0.92) for the BE-with-dysplasia class, and 0.88 (95% CI, 0.84-0.92) for the adenocarcinoma class. The proposed model achieved a mean accuracy of 0.83 (95% CI, 0.80-0.86) and marginally outperformed the sliding window approach on the same testing set. The F1 scores of the attention-based model were at least 8% higher for each class compared with the sliding window approach: 0.68 (95% CI, 0.61-0.75) vs 0.61 (95% CI, 0.53-0.68) for the normal class, 0.72 (95% CI, 0.63-0.80) vs 0.58 (95% CI, 0.45-0.69) for the BE-no-dysplasia class, 0.30 (95% CI, 0.11-0.48) vs 0.22 (95% CI, 0.11-0.33) for the BE-with-dysplasia class, and 0.67 (95% CI, 0.54-0.77) vs 0.58 (95% CI, 0.44-0.70) for the adenocarcinoma class. However, this outperformance was not statistically significant. Conclusions and Relevance: Results of this study suggest that the proposed attention-based deep neural network framework for BE and esophageal adenocarcinoma detection is important because it is based solely on tissue-level annotations, unlike existing methods that are based on regions of interest. This new model is expected to open avenues for applying deep learning to digital pathology.

Automated Detection of Celiac Disease on Duodenal Biopsy Slides: A Deep Learning Approach.

CONTEXT: Celiac disease (CD) prevalence and diagnosis have increased substantially in recent years. The current gold standard for CD confirmation is visual examination of duodenal mucosal biopsies. An accurate computer-aided biopsy analysis system using deep learning can help pathologists diagnose CD more efficiently. SUBJECTS AND METHODS: In this study, we trained a deep learning model to detect CD on duodenal biopsy images. Our model uses a state-of-the-art residual convolutional neural network to evaluate patches of duodenal tissue and then aggregates those predictions for whole-slide classification. We tested the model on an independent set of 212 images and evaluated its classification results against reference standards established by pathologists. RESULTS: Our model identified CD, normal tissue, and nonspecific duodenitis with accuracies of 95.3%, 91.0%, and 89.2%, respectively. The area under the receiver operating characteristic curve was >0.95 for all classes. CONCLUSIONS: We have developed an automated biopsy analysis system that achieves high performance in detecting CD on biopsy slides. Our system can highlight areas of interest and provide preliminary classification of duodenal biopsies before review by pathologists. This technology has great potential for improving the accuracy and efficiency of CD diagnosis.

Pathologist-level classification of histologic patterns on resected lung adenocarcinoma slides with deep neural networks.

Classification of histologic patterns in lung adenocarcinoma is critical for determining tumor grade and treatment for patients. However, this task is often challenging due to the heterogeneous nature of lung adenocarcinoma and the subjective criteria for evaluation. In this study, we propose a deep learning model that automatically classifies the histologic patterns of lung adenocarcinoma on surgical resection slides. Our model uses a convolutional neural network to identify regions of neoplastic cells, then aggregates those classifications to infer predominant and minor histologic patterns for any given whole-slide image. We evaluated our model on an independent set of 143 whole-slide images. It achieved a kappa score of 0.525 and an agreement of 66.6% with three pathologists for classifying the predominant patterns, slightly higher than the inter-pathologist kappa score of 0.485 and agreement of 62.7% on this test set. All evaluation metrics for our model and the three pathologists were within 95% confidence intervals of agreement. If confirmed in clinical practice, our model can assist pathologists in improving classification of lung adenocarcinoma patterns by automatically pre-screening and highlighting cancerous regions prior to review. Our approach can be generalized to any whole-slide image classification task, and code is made publicly available at https://github.com/BMIRDS/deepslide .

Automating the Paris System for urine cytopathology-A hybrid deep-learning and morphometric approach.

BACKGROUND: The Paris System for Urine Cytopathology (the Paris System) has succeeded in making the analysis of liquid-based urine preparations more reproducible. Any algorithm seeking to automate this system must accurately estimate the nuclear-to-cytoplasmic (N:C) ratio and produce a qualitative "atypia score." The authors propose a hybrid deep-learning and morphometric model that reliably automates the Paris System. METHODS: Whole-slide images (WSI) of liquid-based urine cytology specimens were extracted from 51 negative, 60 atypical, 52 suspicious, and 54 positive cases. Morphometric algorithms were applied to decompose images to their component parts; and statistics, including the NC ratio, were tabulated using segmentation algorithms to create organized data structures, dubbed rich information matrices (RIMs). These RIM objects were enhanced using deep-learning algorithms to include qualitative measures. The augmented RIM objects were then used to reconstruct WSIs with filtering criteria and to generate pancellular statistical information. RESULTS: The described system was used to calculate the N:C ratio for all cells, generate object classifications (atypical urothelial cell, squamous cell, crystal, etc), filter the original WSI to remove unwanted objects, rearrange the WSI to an efficient, condensed-grid format, and generate pancellular statistics containing quantitative/qualitative data for every cell in a WSI. In addition to developing novel techniques for managing WSIs, a system capable of automatically tabulating the Paris System criteria also was generated. CONCLUSIONS: A hybrid deep-learning and morphometric algorithm was developed for the analysis of urine cytology specimens that could reliably automate the Paris System and provide many avenues for increasing the efficiency of digital screening for urine WSIs and other cytology preparations.