Recommendations for diabetic macular edema management by retina specialists and large language model-based artificial intelligence platforms.

PURPOSE: To study the role of artificial intelligence (AI) in developing diabetic macular edema (DME) management recommendations by creating and comparing responses to clinicians in hypothetical AI-generated case scenarios. The study also examined whether its joint recommendations followed national DME management guidelines. METHODS: The AI hypothetically generated 50 ocular case scenarios from 25 patients using keywords like age, gender, type, duration and control of diabetes, visual acuity, lens status, retinopathy stage, coexisting ocular and systemic co-morbidities, and DME-related retinal imaging findings. For DME and ocular co-morbidity management, we calculated inter-rater agreements (kappa analysis) separately for clinician responses, AI-platforms, and the "majority clinician response" (the maximum number of identical clinician responses) and "majority AI-platform" (the maximum number of identical AI responses). Treatment recommendations for various situations were compared to the Indian national guidelines. RESULTS: For DME management, clinicians (ĸ=0.6), AI platforms (ĸ=0.58), and the 'majority clinician response' and 'majority AI response' (ĸ=0.69) had moderate to substantial inter-rate agreement. The study showed fair to substantial agreement for ocular co-morbidity management between clinicians (ĸ=0.8), AI platforms (ĸ=0.36), and the 'majority clinician response' and 'majority AI response' (ĸ=0.49). Many of the current study's recommendations and national clinical guidelines agreed and disagreed. When treating center-involving DME with very good visual acuity, lattice degeneration, renal disease, anaemia, and a recent history of cardiovascular disease, there were clear disagreements. CONCLUSION: For the first time, this study recommends DME management using large language model-based generative AI. The study's findings could guide in revising the global DME management guidelines.

Oral Manifestations of Psychotropic Drugs on the Oral Cavity: Observational Study.

AIM: This study aims to detect the prevalence of oral manifestations in patients with psychiatric disorders on psychotropic medications. MATERIALS AND METHODS: A total of 46 patients above the age of 18 years who have been diagnosed with psychiatric illness and under psychotropic medications were included in this study. Thorough case history and oral findings were recorded. Patients with already existing systemic illness and other oral manifestations were excluded from this study. RESULTS: Out of 46 patients, 34 patients presented with oral manifestations such as xerostomia, sialorrhea, geographic tongue, candidiasis, and burning mouth syndrome, secondary to the use of psychotropic medications. The oral manifestations were significantly higher in the patients under antipsychotics (80.0%), selective serotonin reuptake inhibitor (66.7%), antiepileptics (55.6%), antidepressants (44.4%), benzodiazepine (44.4%), and tricyclic antidepressants (13.7%). CONCLUSION: The commonly used psychotropic medications to treat patients with psychiatric illnesses such as selective serotonin reuptake inhibitor, tricyclic antidepressants, antidepressants, and benzodiazepines exhibited several oral manifestations. However, long-term use of these medications seems to cause oral changes. CLINICAL SIGNIFICANCE: Awareness among psychiatrists about oral changes associated with the use of psychotropic medication will assist them to make necessary modifications in the prescriptions. Dental practitioners will be able to recognize these changes early in the course of the condition and provide appropriate treatment.

Comparison of a formatted versus traditional sign out process for physicians in the emergency department.

INTRODUCTION: Frequent interruptions, critically ill patients, and high patient turnover can make Emergency Department (ED) physician transitions of care (TOCs) challenging. Currently, there is no strict format for TOC in the ED. We structured a formatted ED TOC and evaluated the comparative effects from traditional TOC practice on the perceived quality of sign-out among physicians working in the ED. METHODS: We performed a prospective pre/post-interventional study utilizing convenience sampling in an urban community teaching hospital. The primary outcome was perceived quality of sign-out, as evaluated by the incoming physician one-hour after TOC, using the handoff-Clinical Evaluation Exercise (h-CEX) score with a 9-point scale for each category: Organized/Efficient, Communications Skills, Included Pertinent Information, Clinical Judgment, Patient Focused, Setting, and Overall Sign-Out Quality. Additional evaluation of unexpected tasks and errors from TOC w performed. RESULTS: We included 344 patient TOC observed, of which 147 (43%) were formatted interventions while 197 (57%) were standard TOCs. After analysis in a random effects model, statistically significant improvements among resident physicians were seen for the formatted TOC: patient focused (mean difference 0.40), setting (mean difference 1.05), and overall (mean difference 0.68). The rate of unexpected tasks and errors were higher in the standard TOC, though not statistically significant. CONCLUSION: Resident physicians saw improvement in several h-CEX categories with a formatted TOC. Consistent with prior studies, a formatted TOC for emergency medicine should be strongly considered, especially among learners.

Emergency Department Drug Surveillance (EDDS) hospital's urinalysis results compared with expanded re-testing by an independent laboratory, a pilot study.

INTRODUCTION: Most hospital urine toxicology screens detect a fixed, limited set of common substances. These tests are fast and accurate but may miss emerging trends in substance use in the community and clinical acumen alone is insufficient for identifying new substances. METHODS: This prospective cohort study examined de-identified urine specimens obtained from patients visiting the Emergency Department (ED) at Prince George's Hospital Center (PGHC), between October 15, 2019 to November 6, 2019 and tested positive for one or more substances. The Emergency Department Drug Surveillance System (EDDS) collects quarterly exports from de-identified electronic health records (EHRs) containing urinalysis results for drug related ED visits. We performed a feasibility study of a new urine specimen submission by collecting a stratified sample of 151 urine specimens from PGHC ED patients. The specimens were tested for 240 drugs using liquid chromatography-tandem mass spectrometry (LC-MS/MS). This paper presents a comparison between the PGHC and expanded testing results. RESULTS: The expanded urinalysis panel found more cocaine (37% vs. 20%; p < 0.01) and benzodiazepine positives (21% vs. 11%; p < 0.05) than would have been detected by the hospital screen. Additionally, the expanded toxicology panel identified fentanyl in 4-14% of the samples. CONCLUSION: The EHR data submitted to EDDS from the hospital urine toxicology screen correctly identified hospital substance use patterns over the approximate 1 month study period. The expanded testing also uncovered drugs that the hospital might consider adding to their routine screen. EDDS is a feasible system for monitoring and confirming recent substance use trends among ED patients.

A Deep Language Model for Symptom Extraction From Clinical Text and its Application to Extract COVID-19 Symptoms From Social Media.

Patients experience various symptoms when they haveeither acute or chronic diseases or undergo some treatments for diseases. Symptoms are often indicators of the severity of the disease and the need for hospitalization. Symptoms are often described in free text written as clinical notes in the Electronic Health Records (EHR) and are not integrated with other clinical factors for disease prediction and healthcare outcome management. In this research, we propose a novel deep language model to extract patient-reported symptoms from clinical text. The deep language model integrates syntactic and semantic analysis for symptom extraction and identifies the actual symptoms reported by patients and conditional or negation symptoms. The deep language model can extract both complex and straightforward symptom expressions. We used a real-world clinical notes dataset to evaluate our model and demonstrated that our model achieves superior performance compared to three other state-of-the-art symptom extraction models. We extensively analyzed our model to illustrate its effectiveness by examining each component's contribution to the model. Finally, we applied our model on a COVID-19 tweets data set to extract COVID-19 symptoms. The results show that our model can identify all the symptoms suggested by the Center for Disease Control (CDC) ahead of their timeline and many rare symptoms.

Manual Palpation vs. Femoral Arterial Doppler Ultrasound for Comparison of Pulse Check Time During Cardiopulmonary Resuscitation in the Emergency Department: A Pilot Study.

BACKGROUND: Manual palpation (MP) is frequently employed for pulse checks, but studies have shown that trained medical personnel have difficulty accurately identifying pulselessness or return of spontaneous circulation (ROSC) using MP. Any delays in identifying pulselessness can lead to significant delays in starting or resuming high-quality chest compressions. OBJECTIVES: This study explored whether femoral arterial Doppler ultrasound (FADU) decreases pulse check duration during cardiopulmonary resuscitation (CPR) compared with MP among patients in the emergency department (ED) receiving CPR directed by emergency medicine physicians who had received minimal additional didactic ultrasound training. METHODS: We performed a prospective observational cohort study from October 2018 to May 2019 at an urban community ED. Using convenience sampling, we enrolled patients arriving at our ED or who decompensated during their ED stay and received CPR. For continuous data, median (interquartile range [IQR]) were calculated, and medians were compared using Kruskal-Wallis test. RESULTS: Fifty-two eligible patients were enrolled and 135 pulse checks via MP and 35 via FADU were recorded. MP observations had a median (IQR) of 11.00 (7.36-15.48) s, whereas FADU had a median (IQR) of 8.98 (5.45-13.85) s. There was a difference between the two medians of 2.02 s (p = 0.05). CONCLUSIONS: In this study, the use of FADU was superior to MP in achieving shorter pulse check times. Further research is needed to confirm the accuracy of FADU for identifying ROSC as well as to determine whether FADU can improve clinical outcomes.

A Computational Framework to Analyze the Associations Between Symptoms and Cancer Patient Attributes Post Chemotherapy Using EHR Data.

Patients with cancer, such as breast and colorectal cancer, often experience different symptoms post-chemotherapy. The symptoms could be fatigue, gastrointestinal (nausea, vomiting, lack of appetite), psychoneurological symptoms (depressive symptoms, anxiety), or other types. Previous research focused on understanding the symptoms using survey data. In this research, we propose to utilize the data within the Electronic Health Record (EHR). A computational framework is developed to use a natural language processing (NLP) pipeline to extract the clinician-documented symptoms from clinical notes. Then, a patient clustering method is based on the symptom severity levels to group the patient in clusters. The association rule mining is used to analyze the associations between symptoms and patient attributes (smoking history, number of comorbidities, diabetes status, age at diagnosis) in the patient clusters. The results show that the various symptom types and severity levels have different associations between breast and colorectal cancers and different timeframes post-chemotherapy. The results also show that patients with breast or colorectal cancers, who smoke and have severe fatigue, likely have severe gastrointestinal symptoms six months after the chemotherapy. Our framework can be generalized to analyze symptoms or symptom clusters of other chronic diseases where symptom management is critical.

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.

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.

Correlation of history and physical examination with imaging in traumatic near-shore aquatic head and spinal injury.

OBJECTIVE: It remains unclear whether clinicians can rely on specific symptoms and signs to detect or exclude serious head and spinal injury sustained during near-shore aquatic activities. Our study investigated patients' history of present illness (HPI) and physical examination (PE) for their utility in detecting serious head and spinal injury. METHODS: We conducted a multicenter retrospective comparative analysis of adult patients who were transported from the beach in Ocean City, Maryland, to three nearby emergency departments for possible spinal injury from 2006 through 2017. Patients suspected to have any spinal injury from beach activities were eligible. We excluded patients who could not verbalize their symptoms or with insufficient emergency department records. We compared components of each patient's HPI and PE with radiologic evidence of spinal injury. We calculated sensitivity, specificity, and negative and positive likelihood ratios (LRs). RESULTS: We analyzed 278 patients with suspected spinal injury. Midline spinal tenderness was associated with increased likelihood of thoracic (LR+ 2.6) and lumbar spinal fractures (LR+ 3.5). HPI complaints of paralysis (LR+ 13.9) and sensory loss (LR+ 5.8) had strong associations with spinal cord injuries. Weakness found through PE was also associated with spinal cord injury (LR+ 5.3). CONCLUSIONS: We identified several components of the clinical evaluation that had clinically significant association with spinal injuries from beach-related trauma. While prospective studies are needed to confirm our observations, clinicians may consider these high-risk features in patients with beach-related trauma and adjust testing and level of care appropriately.

Attention Mechanism with BERT for Content Annotation and Categorization of Pregnancy-Related Questions on a Community Q&A Site.

In recent years, the social web has been increasingly used for health information seeking, sharing, and subsequent health-related research. Women often use the Internet or social networking sites to seek information related to pregnancy in different stages. They may ask questions about birth control, trying to conceive, labor, or taking care of a newborn or baby. Classifying different types of questions about pregnancy information (e.g., before, during, and after pregnancy) can inform the design of social media and professional websites for pregnancy education and support. This research aims to investigate the attention mechanism built-in or added on top of the BERT model in classifying and annotating the pregnancy-related questions posted on a community Q&A site. We evaluated two BERT-based models and compared them against the traditional machine learning models for question classification. Most importantly, we investigated two attention mechanisms: the built-in self-attention mechanism of BERT and the additional attention layer on top of BERT for relevant term annotation. The classification performance showed that the BERT-based models worked better than the traditional models, and BERT with an additional attention layer can achieve higher overall precision than the basic BERT model. The results also showed that both attention mechanisms work differently on annotating relevant content, and they could serve as feature selection methods for text mining in general.

Synthesis, Characterization and In Vitro Antimicrobial Evaluation of Novel Pyrazolothiazol-4(5H)-one Derivatives.

Antimicrobial screening of several novel pyrazolothiazol-4(5H)-one derivatives (3a-3j) has been performed. Reaction of aromatic aldehydes with aromatic ketones yielded starting chalcones (1a-1j) which have been subsequently reacted with thiosemicarbazide for obtaining N-thiocarbamoylpyrazole derivatives (2a-2j). These were further cyclized to pyrazolothiazol-4(5H)-one derivatives (3a-3j) in the presence of ethylbromoacetate. The structures of newly synthesized compounds were confirmed by FTIR and (1)H NMR and/or MS. The in vitro antimicrobial activity of these compounds was evaluated against Gram positive bacteria, Gram negative bacteria and fungi. Their minimum inhibitory concentration was determined by tube dilution method. The results showed that most of the compounds have promising antimicrobial activity as compared to standard drugs. Among the test compounds, 2-[5(4-chlorophenyl)-3-phenyl-4,5-dihydropyrazol-1-yl]-thiazol-4(5H)-one (3e) was found to show the most potent antimicrobial activity.