Application of unsupervised deep learning algorithms for identification of specific clusters of chronic cough patients from EMR data.

BACKGROUND: Chronic cough affects approximately 10% of adults. The lack of ICD codes for chronic cough makes it challenging to apply supervised learning methods to predict the characteristics of chronic cough patients, thereby requiring the identification of chronic cough patients by other mechanisms. We developed a deep clustering algorithm with auto-encoder embedding (DCAE) to identify clusters of chronic cough patients based on data from a large cohort of 264,146 patients from the Electronic Medical Records (EMR) system. We constructed features using the diagnosis within the EMR, then built a clustering-oriented loss function directly on embedded features of the deep autoencoder to jointly perform feature refinement and cluster assignment. Lastly, we performed statistical analysis on the identified clusters to characterize the chronic cough patients compared to the non-chronic cough patients. RESULTS: The experimental results show that the DCAE model generated three chronic cough clusters and one non-chronic cough patient cluster. We found various diagnoses, medications, and lab tests highly associated with chronic cough patients by comparing the chronic cough cluster with the non-chronic cough cluster. Comparison of chronic cough clusters demonstrated that certain combinations of medications and diagnoses characterize some chronic cough clusters. CONCLUSIONS: To the best of our knowledge, this study is the first to test the potential of unsupervised deep learning methods for chronic cough investigation, which also shows a great advantage over existing algorithms for patient data clustering.

Differences in Health-Related Outcomes and Health Care Resource Utilization in Breast Cancer Survivors With and Without Type 2 Diabetes.

PURPOSE: Up to 74% of breast cancer survivors (BCS) have at least one preexisting comorbid condition, with diabetes (type 2) common. The purpose of this study was to examine differences in health-related outcomes (anemia, neutropenia, and infection) and utilization of health care resources (inpatient, outpatient, and emergency visits) in BCS with and without diabetes. METHODS: In this retrospective cohort study, data were leveraged from the electronic health records of a large health network linked to the Indiana State Cancer Registry. BCS diagnosed between January 2007 and December 2017 and who had received chemotherapy were included. Multivariable logistic regression and generalized linear models were used to determine differences in health outcomes and health care resources. RESULTS: The cohort included 6851 BCS, of whom 1121 (16%) had a diagnosis of diabetes. BCS were, on average, 55 (standard deviation: 11.88) years old, the majority self-reported race as White (90%), and 48.8% had stage II breast cancer. BCS with diabetes were significantly older (mean age of 60.6 [SD: 10.34] years) than those without diabetes and were often obese (66% had body mass index of ≥33). BCS with diabetes had higher odds of anemia (odds ratio: 1.43; 95% CI: 1.04, 1.96) and infection (odds ratio: 1.86; 95% CI: 1.35, 2.55) and utilized more outpatient resources (P<0.0001). CONCLUSIONS: Diabetes has a deleterious effect on health-related outcomes and health care resource utilization among BCS. These findings support the need for clinical practice guidelines to help clinicians manage diabetes among BCS throughout the cancer trajectory and for coordinated models of care to reduce high resource utilization.

Applying interpretable deep learning models to identify chronic cough patients using EHR data.

BACKGROUND AND OBJECTIVE: Chronic cough (CC) affects approximately 10% of adults. Many disease states are associated with chronic cough, such as asthma, upper airway cough syndrome, bronchitis, and gastroesophageal reflux disease. The lack of an ICD code specific for chronic cough makes it challenging to identify such patients from electronic health records (EHRs). For clinical and research purposes, computational methods using EHR data are urgently needed to identify chronic cough cases. This research aims to investigate the data representations and deep learning algorithms for chronic cough prediction. METHODS: Utilizing real-world EHR data from a large academic healthcare system from October 2005 to September 2015, we investigated Natural Language Representation of the EHR data and systematically evaluated deep learning and traditional machine learning models to predict chronic cough patients. We built these machine learning models using structured data (medication and diagnosis) and unstructured data (clinical notes). RESULTS: The sensitivity and specificity of a transformer-based deep learning algorithm, specifically BERT with attention model, was 0.856 and 0.866, respectively, using structured data (medication and diagnosis). Sensitivity and specificity improved to 0.952 and 0.930 when we combined structured data with symptoms extracted from clinical notes. We further found that the attention mechanism of deep learning models can be used to extract important features that drive the prediction decisions. Compared with our previously published rule-based algorithm, the deep learning algorithm can identify more chronic cough patients with structured data. CONCLUSIONS: By applying deep learning models, chronic cough patients can be reliably identified for prospective or retrospective research through medication and diagnosis data, widely available in EHR and electronic claims data, thus improving the generalizability of the patient identification algorithm. Deep learning models can identify chronic cough patients with even higher sensitivity and specificity when structured and unstructured EHR data are utilized. We anticipate language-based data representation and deep learning models developed in this research could also be productively used for other disease prediction and case identification.

Rare case of rapidly reaccumulating ascites.

Eosinophilic ascites is a rare type of exudative ascites most commonly caused by eosinophilic gastroenteritis. Here, a 57-year-old man presents with sudden-onset abdominal distension associated with nausea, vomiting and decreased appetite for 10 days. Physical examination revealed significant abdominal distention and fluid wave. Initial labs showed leucocytosis and mild peripheral eosinophilia. Imaging of his abdomen revealed severe ascites, no features of cirrhosis and diffuse inflammatory changes involving the jejunum and ileum. Diagnostic paracentesis showed exudative, ascitic fluid with predominant eosinophilia. Cytology of the ascitic fluid and blind biopsies taken during oesophagogastroduodenoscopy and enteroscopy were both negative for malignancy. The ascites reaccumulated rapidly, requiring five rounds of large-volume paracentesis during hospitalisation. Empiric treatment for suspected eosinophilic gastroenteritis with intravenous steroids improved and stabilised the patient's ascites for discharge. Parasitic workup resulted positively for Toxocara antibodies on ELISA. On 2-week outpatient follow-up, a course of albendazole resolved all gastrointestinal symptoms.

Analyzing the symptoms in colorectal and breast cancer patients with or without type 2 diabetes using EHR data.

This research extracted patient-reported symptoms from free-text EHR notes of colorectal and breast cancer patients and studied the correlation of the symptoms with comorbid type 2 diabetes, race, and smoking status. An NLP framework was developed first to use UMLS MetaMap to extract all symptom terms from the 366,398 EHR clinical notes of 1694 colorectal cancer (CRC) patients and 3458 breast cancer (BC) patients. Semantic analysis and clustering algorithms were then developed to categorize all the relevant symptoms into eight symptom clusters defined by seed terms. After all the relevant symptoms were extracted from the EHR clinical notes, the frequency of the symptoms reported from colorectal cancer (CRC) and breast cancer (BC) patients over three time-periods post-chemotherapy was calculated. Logistic regression (LR) was performed with each symptom cluster as the response variable while controlling for diabetes, race, and smoking status. The results show that the CRC and BC patients with Type 2 Diabetes (T2D) were more likely to report symptoms than CRC and BC without T2D over three time-periods in the cancer trajectory. We also found that current smokers were more likely to report anxiety (CRC, BC), neuropathic symptoms (CRC, BC), anxiety (BC), and depression (BC) than non-smokers.

Association of Comorbid Diabetes With Clinical Outcomes and Healthcare Utilization in Colorectal Cancer Survivors.

OBJECTIVES: To compare clinical outcomes and healthcare utilization in colorectal cancer (CRC) survivors with and without diabetes. SAMPLE & SETTING: CRC survivors (N = 3,287) were identified from a statewide electronic health record database using International Classification of Diseases (ICD) codes. Data were extracted on adults aged 21 years or older with an initial diagnosis of stage II or III CRC with diabetes present before CRC diagnosis or no diagnosis of diabetes (control). METHODS & VARIABLES: ICD codes were used to extract diabetes diagnosis and clinical outcome variables. Healthcare utilization was determined by encounter type. Data were analyzed using descriptive statistics, multivariable logistic, and Cox regression. RESULTS: CRC survivors with diabetes were more likely to develop anemia and infection than CRC survivors without diabetes. In addition, CRC survivors with diabetes were more likely to utilize emergency resources sooner than CRC survivors without diabetes. IMPLICATIONS FOR NURSING: Oncology nurses can facilitate the early identification of high-risk survivor groups, reducing negative clinical outcomes and unnecessarily high healthcare resource utilization in CRC survivors with diabetes.