Critical Risk Assessment, Diagnosis, and Survival Analysis of Breast Cancer.

Breast cancer is the most prevalent type of cancer in women. Risk factor assessment can aid in directing counseling regarding risk reduction and breast cancer surveillance. This research aims to (1) investigate the relationship between various risk factors and breast cancer incidence using the BCSC (Breast Cancer Surveillance Consortium) Risk Factor Dataset and create a prediction model for assessing the risk of developing breast cancer; (2) diagnose breast cancer using the Breast Cancer Wisconsin diagnostic dataset; and (3) analyze breast cancer survivability using the SEER (Surveillance, Epidemiology, and End Results) Breast Cancer Dataset. Applying resampling techniques on the training dataset before using various machine learning techniques can affect the performance of the classifiers. The three breast cancer datasets were examined using a variety of pre-processing approaches and classification models to assess their performance in terms of accuracy, precision, F-1 scores, etc. The PCA (principal component analysis) and resampling strategies produced remarkable results. For the BCSC Dataset, the Random Forest algorithm exhibited the best performance out of the applied classifiers, with an accuracy of 87.53%. Out of the different resampling techniques applied to the training dataset for training the Random Forest classifier, the Tomek Link exhibited the best test accuracy, at 87.47%. We compared all the models used with previously used techniques. After applying the resampling techniques, the accuracy scores of the test data decreased even if the training data accuracy increased. For the Breast Cancer Wisconsin diagnostic dataset, the K-Nearest Neighbor algorithm had the best accuracy with the original dataset test set, at 94.71%, and the PCA dataset test set exhibited 95.29% accuracy for detecting breast cancer. Using the SEER Dataset, this study also explores survival analysis, employing supervised and unsupervised learning approaches to offer insights into the variables affecting breast cancer survivability. This study emphasizes the significance of individualized approaches in the management and treatment of breast cancer by incorporating phenotypic variations and recognizing the heterogeneity of the disease. Through data-driven insights and advanced machine learning, this study contributes significantly to the ongoing efforts in breast cancer research, diagnostics, and personalized medicine.

Texture-Based Classification to Overcome Uncertainty between COVID-19 and Viral Pneumonia Using Machine Learning and Deep Learning Techniques.

The SARS-CoV-2 virus, responsible for COVID-19, often manifests symptoms akin to viral pneumonia, complicating early detection and potentially leading to severe COVID pneumonia and long-term effects. Particularly affecting young individuals, the elderly, and those with weakened immune systems, the accurate classification of COVID-19 poses challenges, especially with highly dimensional image data. Past studies have faced limitations due to simplistic algorithms and small, biased datasets, yielding inaccurate results. In response, our study introduces a novel classification model that integrates advanced texture feature extraction methods, including GLCM, GLDM, and wavelet transform, within a deep learning framework. This innovative approach enables the effective classification of chest X-ray images into normal, COVID-19, and viral pneumonia categories, overcoming the limitations encountered in previous studies. Leveraging the unique textures inherent to each dataset class, our model achieves superior classification performance, even amidst the complexity and diversity of the data. Moreover, we present comprehensive numerical findings demonstrating the superiority of our approach over traditional methods. The numerical results highlight the accuracy (random forest (RF): 0.85; SVM (support vector machine): 0.70; deep learning neural network (DLNN): 0.92), recall (RF: 0.85, SVM: 0.74, DLNN: 0.93), precision (RF: 0.86, SVM: 0.71, DLNN: 0.87), and F1-Score (RF: 0.86, SVM: 0.72, DLNN: 0.89) of our proposed model. Our study represents a significant advancement in AI-based diagnostic systems for COVID-19 and pneumonia, promising improved patient outcomes and healthcare management strategies.

Correlation Aware Relevance-Based Semantic Index for Clinical Big Data Repository.

In this paper, we focus on indexing mechanisms for unstructured clinical big integrated data repository systems. Clinical data is unstructured and heterogeneous, which comes in different files and formats. Accessing data efficiently and effectively are critical challenges. Traditional indexing mechanisms are difficult to apply on unstructured data, especially by identifying correlation information between clinical data elements. In this research work, we developed a correlation-aware relevance-based index that retrieves clinical data by fetching most relevant cases efficiently. In our previous work, we designed a methodology that categorizes medical data based on the semantics of data elements and merges them into an integrated repository. We developed a data integration system for medical data sources that combines heterogeneous medical data and provides access to knowledge-based database repositories to different users. In this research work, we designed an indexing system using semantic tags extracted from clinical data sources and medical ontologies that retrieves relevant data from database repositories and speeds up the process of data retrieval. Our objective is to provide an integrated biomedical database repository that can be used by radiologists as a reference, or for patient care, or by researchers. In this paper, we focus on designing a technique that performs data processing for data integration, learn the semantic properties of data elements, and develop a correlation-aware topic index that facilitates efficient data retrieval. We generated semantic tags by identifying key elements from integrated clinical cases using topic modeling techniques. We investigated a technique that identifies tags for merged categories and provides an index to fetch data from an integrated database repository. We developed a topic coherence matrix that shows how well a topic is supported by a corpus from clinical cases and medical ontologies. We were able to find more relevant results using an annotation index from an integrated database repository, and there was a 61% increase in a recall. We evaluated results with the help of experts and compared them with naive index (index with all terms from the corpus). Our approach improved data retrieval quality by providing most relevant results and reduced data retrieval time as we applied correlation-aware index on an integrated data repository. Topic indexing approach proposed in this research work identifies tags based on a correlation between different data elements, improves data retrieval time, and provides most relevant cases as an outcome of this system.

Geographic Dialysis Facility Density and Early Dialysis Initiation.

Importance: The decision of when to start maintenance hemodialysis may be affected by health system-level support for high-intensity care as manifested by area dialysis facility density. Yet an association between early hemodialysis initiation and higher area density of dialysis facilities has not been shown. Objective: To examine whether there is an association between area dialysis facility density and earlier dialysis initiation. Design, Setting, and Participants: Cross-sectional analysis was conducted of publicly reported claims and geographic-based population data collected in the Medical Evidence files of the US Renal Data System (USRDS), a comprehensive registry of all patients initiating hemodialysis in the US, from calendar years 2011 through 2019. Data were linked to the American Community Survey, using residential zip codes, and then to health service area (HSA) primary care and hospitalization benchmarks, using the Dartmouth Atlas crosswalk. Data were analyzed from November 1, 2021, to August 31, 2023. Exposure: Dialysis facility density at the level of HSA (number of dialysis facilities per 100 000 HSA residents) split into 5 categories. Main Outcomes and Measures: The odds of hemodialysis initiation at an estimated glomerular filtration rate (eGFR) greater than 10 mL/min/1.73 m2 vs less than or equal to 10 mL/min/1.73 m2. Results: Hemodialysis was initiated in a total of 844 466 individuals at 3397 HSAs at a mean (SD) eGFR of 8.9 (3.8) mL/min/1.73 m2. Their mean (SD) age was 63.5 (14.7) years, and 484 346 participants (57.4%) were men. In the HSA category with the highest facility density, individuals were younger (63.3 vs 65.2 years in least-dense HSAs), poorer (mean percent of households living in poverty, 10.4% vs 8.4%), and more commonly had a higher percentage of Black individuals (40.6% vs 11.3%). More individuals in the dialysis-dense HSAs than least-dense HSAs had diabetes (60.1% vs 58.5%) and fewer had access to predialysis nephrology care (60.8% vs 64.1%); the rates of heart failure and immobility varied, but not in a consistent pattern, by HSA dialysis density. The mean (SD) facility density was 4.1 (1.89) centers per 100 000 population in the most dialysis-dense HSAs. Compared with patients in HSAs with a mean of 1.0 per 100 000 population, the odds of hemodialysis initiation at eGFR greater than 10 mL/min/1.73 m2 were 1.07 (95% CI, 1.03-1.11) for patients in the densest HSAs, and compared with HSAs with 0 facilities, the odds of early hemodialysis initiation were 1.06 (95% CI, 1.02-1.10) for patients in the densest HSAs. Conclusions and Relevance: In this cross-sectional study of USRDS- and HSA-level data, HSA dialysis density was associated with early hemodialysis initiation.

A Rare Case of Co-existing Non-small Cell Lung Carcinoma and Non-tuberculous Mycobacteria.

A solitary pulmonary mass is commonly associated with malignancy; however, the possibility of co-existence with a pulmonary infection is rarely considered. Here, we present an extraordinary case, underscoring the importance of considering the possibility of concurrent lung cancer even when a bronchoscopy examination and bronchial lavage yield a positive mycobacterium culture result.

Biomedical heterogeneous data categorization and schema mapping toward data integration.

Data integration is a well-motivated problem in the clinical data science domain. Availability of patient data, reference clinical cases, and datasets for research have the potential to advance the healthcare industry. However, the unstructured (text, audio, or video data) and heterogeneous nature of the data, the variety of data standards and formats, and patient privacy constraint make data interoperability and integration a challenge. The clinical text is further categorized into different semantic groups and may be stored in different files and formats. Even the same organization may store cases in different data structures, making data integration more challenging. With such inherent complexity, domain experts and domain knowledge are often necessary to perform data integration. However, expert human labor is time and cost prohibitive. To overcome the variability in the structure, format, and content of the different data sources, we map the text into common categories and compute similarity within those. In this paper, we present a method to categorize and merge clinical data by considering the underlying semantics behind the cases and use reference information about the cases to perform data integration. Evaluation shows that we were able to merge 88% of clinical data from five different sources.

Risk Amplifiers for Vascular Disease and CKD in South Asians: When Intrinsic ß-Cell Dysfunction Meets a High-Carbohydrate Diet.

South Asians, comprising almost one fourth of the world population, are at higher risk of type 2 diabetes mellitus, hypertension, cardiovascular disease, and CKD compared with other ethnic groups. This has major public health implications in South Asia and in other parts of the world to where South Asians have immigrated. The interplay of various modifiable and nonmodifiable risk factors confers this risk. Traditional models of cardiometabolic disease progression and CKD evaluation may not be applicable in this population with a unique genetic predisposition and phenotype. A wider understanding of dietary and lifestyle influences, genetic and metabolic risk factors, and the pitfalls of conventional equations estimating kidney function in this population are required in providing care for kidney diseases. Targeted screening of this population for metabolic and vascular risk factors and individualized management plan for disease management may be necessary. Addressing unhealthy dietary patterns, promoting physical activity, and medication management that adheres to cultural factors are crucial steps to mitigate the risk of cardiovascular disease and CKD in this population. In South Asian countries, a large rural and urban community-based multipronged approach using polypills and community health workers to decrease the incidence of these diseases may be cost-effective.

Shorter versus longer corticosteroid duration and recurrent immune checkpoint inhibitor-associated AKI.

BACKGROUND: Corticosteroids are the mainstay of treatment for immune checkpoint inhibitor-associated acute kidney injury (ICPi-AKI), but the optimal duration of therapy has not been established. Prolonged use of corticosteroids can cause numerous adverse effects and may decrease progression-free survival among patients treated with ICPis. We sought to determine whether a shorter duration of corticosteroids was equally efficacious and safe as compared with a longer duration. METHODS: We used data from an international multicenter cohort study of patients diagnosed with ICPi-AKI from 29 centers across nine countries. We examined whether a shorter duration of corticosteroids (28 days or less) was associated with a higher rate of recurrent ICPi-AKI or death within 30 days following completion of corticosteroid treatment as compared with a longer duration (29-84 days). RESULTS: Of 165 patients treated with corticosteroids, 56 (34%) received a shorter duration of treatment and 109 (66%) received a longer duration. Patients in the shorter versus longer duration groups were similar with respect to baseline and ICPi-AKI characteristics. Five of 56 patients (8.9%) in the shorter duration group and 12 of 109 (11%) in the longer duration group developed recurrent ICPi-AKI or died (p=0.90). Nadir serum creatinine in the first 14, 28, and 90 days following completion of corticosteroid treatment was similar between groups (p=0.40, p=0.56, and p=0.89, respectively). CONCLUSION: A shorter duration of corticosteroids (28 days or less) may be safe for patients with ICPi-AKI. However, the findings may be susceptible to unmeasured confounding and further research from randomized clinical trials is needed.

Polycystic Kidney/Liver Disease.

Autosomal dominant polycystic kidney disease (ADPKD) is a genetic disorder that leads to chronic kidney disease and end-stage kidney disease (ESKD). Polycystic liver disease (PCLD) is the most common extrarenal manifestation of ADPKD. Though isolated PCLD and PCLD due to ADPKD are genetically distinct, they follow a similar clinical course of hepatomegaly from multiple cysts with preserved liver function. Tolvaptan use in ADPKD can slow down the deterioration of renal function and growth of cysts. Somatostatin analogs can slow the growth of polycystic livers but the effect is short-lived. The only curative therapy for PCLD is liver transplantation. Renal transplantation can significantly improve survival in patients with ESKD due to ADPKD.

Test-Retest Reliability and Internal Consistency of the Coronary Artery Disease Education Questionnaire Short Version in the Marathi Language.

Introduction Patient education specific to the disease must be incorporated into the management of coronary artery disease (CAD) or any other disease as the patients understand the education provided in their native language better. The brief version of the CAD education questionnaire is a valid and reliable instrument for evaluating patients' understanding of the condition. To the best of our knowledge and understanding from the literature review, no questionnaire evaluating the knowledge of CAD in the Marathi language had been found because of which this study was carried out. Methods For the process of translation and cross-cultural adaptation, the framework for self-report measures was taken into consideration in which qualified translators translated both ways forward into the Marathi language and backward into the English language. The translators and a recording observer combined their efforts to create one synthesized version. The questionnaire was fine-tuned by the expert group to produce the final version and 30 diagnosed cases of CAD were tested with the pre-final version. The Marathi version of the questionnaire's validity and reliability were evaluated using Cohen's kappa (k) and Cronbach's alpha (α). Results Thirty individuals with CAD were recruited (mean age 68±12.36, consisting of 22 males and 08 females) to test the pre-final version, and equivalence was tested for every item by probing the participants for the understanding of the item. The Likert scale demonstrated that patients understood the purpose of each question. A total of 200 participants - 153 males population and 47 females with a mean age of 66.64±5.6094 years who can read and speak in the Marathi language were considered to assess the test-retest reliability and internal consistency who completed the questionnaire twice, with a gap of two weeks but only 188 participant's data was analyzed as twelve participants dropped out of the study because they could not report due to transportation and health-related issues. The obtained α value demonstrated satisfactory internal consistency, while the k value indicated almost perfect agreement. Conclusion The study concluded that the Marathi version of the CAD education questionnaire short version is reliable and cross-culturally adapted; therefore, it is an effective tool for evaluating the knowledge of CAD among Marathi language-speaking patients.

Genitourinary Cancer: Updates on Treatments and Their Impact on the Kidney.

Genitourinary cancers are diverse in their presentation, prevalence, and mortality risk. Although there have been significant advancements in medical (eg, immune checkpoint inhibitors and tyrosine kinase inhibitors) and surgical treatments of genitourinary cancers, patients are still at risk for chronic kidney disease, hypertension, and electrolyte derangements in the short and long term. In addition, pre-existing kidney disease may increase the risk of developing some genitourinary cancers. This review focuses on the kidney-related effects of treatments for renal cell carcinoma and bladder and prostate cancers.

Impact of technology-based interventions on linking potential kidney donors and transplant candidates: a scoping review.

Despite the demonstrated survival advantage in end-stage kidney disease (ESKD) patients of a preemptive living donor kidney transplantation (LDKT), there has been a decline in LDKT among African American and Hispanic populations. We performed a scoping review and summarized the evidence about the use of technology-based interventions (TBI) to not only increase knowledge and awareness of LDKT but also link living donors with transplant candidates. We evaluated 31 studies and characterized them into "transplant-candidate facing" TBI, "transplant donor facing" TBI, and "interactive websites" targeting both donors and candidates. For the patient-facing interventions, 60% of studies suggested an increased likelihood of linking possible donors and candidates. The donor-facing interventions showed an increase in donor awareness and 75% of these interventions suggested increasing donor-candidate linkage. This study also demonstrates that TBI (regardless of medium) that are accessible and customized to the specific target population can potentially increase linkage of donors to recipients and serve as effective guides to connect potential donors to transplant candidates.

Acute kidney injury in patients treated with immune checkpoint inhibitors.

BACKGROUND: Immune checkpoint inhibitor-associated acute kidney injury (ICPi-AKI) has emerged as an important toxicity among patients with cancer. METHODS: We collected data on 429 patients with ICPi-AKI and 429 control patients who received ICPis contemporaneously but who did not develop ICPi-AKI from 30 sites in 10 countries. Multivariable logistic regression was used to identify predictors of ICPi-AKI and its recovery. A multivariable Cox model was used to estimate the effect of ICPi rechallenge versus no rechallenge on survival following ICPi-AKI. RESULTS: ICPi-AKI occurred at a median of 16 weeks (IQR 8-32) following ICPi initiation. Lower baseline estimated glomerular filtration rate, proton pump inhibitor (PPI) use, and extrarenal immune-related adverse events (irAEs) were each associated with a higher risk of ICPi-AKI. Acute tubulointerstitial nephritis was the most common lesion on kidney biopsy (125/151 biopsied patients [82.7%]). Renal recovery occurred in 276 patients (64.3%) at a median of 7 weeks (IQR 3-10) following ICPi-AKI. Treatment with corticosteroids within 14 days following ICPi-AKI diagnosis was associated with higher odds of renal recovery (adjusted OR 2.64; 95% CI 1.58 to 4.41). Among patients treated with corticosteroids, early initiation of corticosteroids (within 3 days of ICPi-AKI) was associated with a higher odds of renal recovery compared with later initiation (more than 3 days following ICPi-AKI) (adjusted OR 2.09; 95% CI 1.16 to 3.79). Of 121 patients rechallenged, 20 (16.5%) developed recurrent ICPi-AKI. There was no difference in survival among patients rechallenged versus those not rechallenged following ICPi-AKI. CONCLUSIONS: Patients who developed ICPi-AKI were more likely to have impaired renal function at baseline, use a PPI, and have extrarenal irAEs. Two-thirds of patients had renal recovery following ICPi-AKI. Treatment with corticosteroids was associated with improved renal recovery.

A Systematic Review of Immune Checkpoint Inhibitor-Associated Glomerular Disease.

INTRODUCTION: Immune checkpoint inhibitors (ICIs) are increasingly used to treat cancers. Kidney immune-related adverse events (IRAEs) are now well recognized, with the incidence of IRAEs ranging from 2% to 5%. Most of the initial data related to kidney IRAEs have focused on acute interstitial nephritis (AIN). There are minimal data on the types and relative frequencies of glomerular diseases associated with ICIs, their treatment, and outcomes. METHODS: We performed a systematic review and meta-analysis of all biopsy-proven published cases/series of glomerular pathology associated with ICIs. We searched the MEDLINE, EMBASE, and Cochrane databases from inception to February 2020. We abstracted patient-level data, including demographics, cancer and ICI therapy details, and characteristics of kidney injury. RESULTS: After screening, 27 articles with 45 cases of biopsy-confirmed ICI-associated glomerular disease were identified. Several lesion types were observed, with the most frequent being pauci-immune glomerulonephritis (GN) and renal vasculitis (27%), podocytopathies (24%), and complement 3 GN (C3GN; 11%). Concomitant AIN was reported in 41%. Most patients had ICIs discontinued (88%), and nearly all received corticosteroid treatment (98%). Renal replacement therapy (RRT) was required in 25%. Most patients had full (31%) or partial (42%) recovery from acute kidney injury (AKI), although 19% remained dialysis-dependent, and approximately one-third died. Complete or partial remission of proteinuria was achieved in 45% and 38%, respectively. CONCLUSION: Multiple forms of ICI-associated glomerular disease have been described. Pauci-immune GN, podocytopathies, and C3GN are the most frequently reported lesions. ICI-associated glomerular disease may be associated with poor kidney and mortality outcomes. Oncologists and nephrologists must be aware of glomerular pathologies associated with ICIs and consider obtaining a kidney biopsy specimen when features atypical for AIN are present.

Renal Cell Cancer and Chronic Kidney Disease.

The association between chronic kidney disease (CKD) and renal cell carcinoma (RCC) is bidirectional and multifactorial. Risk factors such as hypertension, diabetes mellitus, obesity, and smoking increase the risk of both CKD and RCC. CKD can lead to RCC via an underlying cystic disease or oxidative stress. RCC can cause CKD because of the tumor itself, surgical reduction of renal mass (either partial or radical nephrectomy), and perioperative acute kidney injury. Medical therapies such as immune checkpoint inhibitors and vascular endothelial growth factor inhibitors can lead to acute kidney injury and resultant CKD. Clinicians need to be aware of the complex, bidirectional interplay between both diseases.

Chronic Kidney Disease in Cancer Survivors.

As breakthroughs in cancer care are leading to improved long-term outcomes in a subset of advanced cancers, there is a growing population of long-term cancer survivors that are at risk of long-term complications. In this review, we summarize what is known about chronic kidney disease in cancer survivors, focusing on the following high-risk groups: survivors of childhood cancers, stem cell transplant recipients, patients with renal cell carcinoma, patients exposed to cisplatin and other nephrotoxic chemotherapies, and patients receiving immunotherapy for cancer. As new anticancer therapies are developed, more research is needed to understand the long-term risks of kidney function decline and to devise methods to prevent chronic kidney disease.

Phase Ib Dose-escalation/Expansion Trial of Ribociclib in Combination With Everolimus and Exemestane in Postmenopausal Women with HR+, HER2- Advanced Breast Cancer.

PURPOSE: Report results of the phase Ib dose-escalation/expansion study of triplet therapy with cyclin-dependent kinases 4 and 6 (CDK4/6) inhibitor (ribociclib), mTOR inhibitor (everolimus), and endocrine therapy (exemestane). PATIENTS AND METHODS: Postmenopausal women with hormone receptor-positive (HR+), human epidermal growth factor receptor 2-negative (HER2-), pretreated, advanced breast cancer (ABC) were enrolled. The primary objective of the dose-escalation phase was to estimate the MTD and recommended phase II dose (RP2D) of triplet therapy through evaluation of the incidence of dose-limiting toxicities. Safety, tolerability, and efficacy of the RP2D were evaluated in the dose-expansion phase in patients naïve or refractory to CDK4/6 inhibitor therapy. RESULTS: Patients (N = 116) received triplet therapy (n = 83 in the dose-escalation phase; n = 33 in the dose-expansion phase). A dose-dependent drug-drug interaction was observed for everolimus, with exposure increasing two- to fourfold in the presence of ribociclib. The RP2D was determined to be ribociclib 300 mg once daily, 3 weeks on/1 week off in a 4-week cycle, plus everolimus 2.5 mg once daily, plus exemestane 25 mg once daily taken with food. The safety profile was consistent with the known profiles of the combination partners, and preliminary evidence of antitumor activity was observed. Higher ESR1 gene expression trended with better treatment response to triplet therapy; higher gene expression of MAPK pathway genes trended with worse treatment response. CONCLUSIONS: Triplet therapy with endocrine therapy and mTOR and CDK4/6 inhibition provides clinical benefit and an acceptable safety profile in previously treated postmenopausal women with HR+, HER2- ABC.

Acute Kidney Injury in Hospitalized Patients with COVID-19.

IMPORTANCE: Preliminary reports indicate that acute kidney injury (AKI) is common in coronavirus disease (COVID)-19 patients and is associated with worse outcomes. AKI in hospitalized COVID-19 patients in the United States is not well-described. OBJECTIVE: To provide information about frequency, outcomes and recovery associated with AKI and dialysis in hospitalized COVID-19 patients. DESIGN: Observational, retrospective study. SETTING: Admitted to hospital between February 27 and April 15, 2020. PARTICIPANTS: Patients aged ≥18 years with laboratory confirmed COVID-19 Exposures: AKI (peak serum creatinine increase of 0.3 mg/dL or 50% above baseline). Main Outcomes and Measures: Frequency of AKI and dialysis requirement, AKI recovery, and adjusted odds ratios (aOR) with mortality. We also trained and tested a machine learning model for predicting dialysis requirement with independent validation. RESULTS: A total of 3,235 hospitalized patients were diagnosed with COVID-19. AKI occurred in 1406 (46%) patients overall and 280 (20%) with AKI required renal replacement therapy. The incidence of AKI (admission plus new cases) in patients admitted to the intensive care unit was 68% (553 of 815). In the entire cohort, the proportion with stages 1, 2, and 3 AKI were 35%, 20%, 45%, respectively. In those needing intensive care, the respective proportions were 20%, 17%, 63%, and 34% received acute renal replacement therapy. Independent predictors of severe AKI were chronic kidney disease, systolic blood pressure, and potassium at baseline. In-hospital mortality in patients with AKI was 41% overall and 52% in intensive care. The aOR for mortality associated with AKI was 9.6 (95% CI 7.4-12.3) overall and 20.9 (95% CI 11.7-37.3) in patients receiving intensive care. 56% of patients with AKI who were discharged alive recovered kidney function back to baseline. The area under the curve (AUC) for the machine learned predictive model using baseline features for dialysis requirement was 0.79 in a validation test. CONCLUSIONS AND RELEVANCE: AKI is common in patients hospitalized with COVID-19, associated with worse mortality, and the majority of patients that survive do not recover kidney function. A machine-learned model using admission features had good performance for dialysis prediction and could be used for resource allocation.

Ontology-Based Radiology Teaching File Summarization, Coverage, and Integration.

Radiology teaching file repositories contain a large amount of information about patient health and radiologist interpretation of medical findings. Although valuable for radiology education, the use of teaching file repositories has been hindered by the ability to perform advanced searches on these repositories given the unstructured format of the data and the sparseness of the different repositories. Our term coverage analysis of two major medical ontologies, Radiology Lexicon (RadLex) and Unified Medical Language System (UMLS) Systematized Nomenclature of Medicine Clinical Terms (SNOMED CT), and two teaching file repositories, Medical Imaging Resource Community (MIRC) and MyPacs, showed that both ontologies combined cover 56.3% of terms in the MIRC and only 17.9% of terms in MyPacs. Furthermore, the overlap between the two ontologies (i.e., terms included by both the RadLex and UMLS SNOMED CT) was a mere 5.6% for the MIRC and 2% for the RadLex. Clustering the content of the teaching file repositories showed that they focus on different diagnostic areas within radiology. The MIRC teaching file covers mostly pediatric cases; a few cases are female patients with heart-, chest-, and bone-related diseases. The MyPacs contains a range of different diseases with no focus on a particular disease category, gender, or age group. MyPacs also provides a wide variety of cases related to the neck, face, heart, chest, and breast. These findings provide valuable insights on what new cases should be added or how existent cases may be integrated to provide more comprehensive data repositories. Similarly, the low-term coverage by the ontologies shows the need to expand ontologies with new terminology such as new terms learned from these teaching file repositories and validated by experts. While our methodology to organize and index data using clustering approaches and medical ontologies is applied to teaching file repositories, it can be applied to any other medical clinical data.