In this paper, we investigate the viability of using a machine learning (ML) model to predict whether an individual will suffer cardiovascular issues as a result of long COVID. Prior research has already been conducted, but those studies focused on either the detection of long COVID using prior condition or prediction of long term cardiovascular effects using COVID symptoms/condition. Our research has been primarily concerned with creating a method by which everyday people could have their risk of long COVID cardiovascular complication assessed. As such, we attempted to develop a model that could predict the occurrence of palpitations using accessible markers of pre-COVID condition (age and body mass index (BMI)). In order to balance the dataset, SMOTE was used to oversample those who experienced palpitations to 30% of the overall data. Eight different models were addressed, with the most promising, XGBoost, further refined via hyperparameter optimization. Its tuning was first done solely focused on accuracy. After it appeared that accuracy had been maximized, it was then tuned to reduce validation loss. Analysis suggested that heart palpitations could be predicted to a high degree of accuracy using only age and body mass index (BMI) with an ensemble tree model. A consistently high validation loss, however, indicated that the model would benefit from training on more robust datasets.

The Chief Complaint (CC) is a crucial component of a patient’s medical record as it describes the main reason or concern for seeking medical care. It provides critical information for healthcare providers to make informed decisions about patient care. However, documenting CCs can be time-consuming for healthcare providers, especially in busy emergency departments. To address this issue, an autocompletion tool that suggests accurate and well-formatted phrases or sentences for clinical notes can be a valuable resource for triage nurses. In this study, we utilized text generation techniques to develop machine learning models using CC data. In our proposed work, we train a Long Short-Term Memory (LSTM) model and fine-tune three different variants of Biomedical Generative Pretrained Transformers (BioGPT), namely microsoft/biogpt, microsoft/BioGPT-Large, and microsoft/BioGPT-Large-PubMedQA. Additionally, we tune a prompt by incorporating exemplar CC sentences, utilizing the OpenAI API of GPT-4. We evaluate the models’ performance based on the perplexity score, modified BERTScore, and cosine similarity score. The results show that BioGPT-Large exhibits superior performance compared to the other models. It consistently achieves a remarkably low perplexity score of 1.65 when generating CC, whereas the baseline LSTM model achieves the best perplexity score of 170. Further, we evaluate and assess the proposed models’ performance and the outcome of GPT-4.0. Our study demonstrates that utilizing LLMs such as BioGPT, leads to the development of an effective autocompletion tool for generating CC documentation in healthcare settings.

The dropout rate among minority students towards the end of their higher education studies is a significant issue. This research work centers on the analysis of student success at Marquette University, WI, USA, and holds significant importance from both an academic achievement perspective and for the nation’s growth. The Institute of Woman’s Leadership at Marquette University in collaboration with EECE faculty is investigating the cause of retention and dropout rates of minority students and preparing to come up with a solution to prevent that. We possess two datasets of 136k undergraduate and 29k PhD/graduate students. The main object of this study was to analyze the success rate of students based on their academic background and demographic data and to investigate how factors such as gender, race, and previous academic performance impact the success rate of minority students in higher education. Another important goal is to assess the impact of the Covid-19 pandemic on the success rate of these students. Several different classification algorithms Decision Tree, Random Forest, Logistic Regression, etc. were applied to the resampled datasets for predicting student success. Out of all the classifiers applied Random Forest had the best accuracy at 96.7% for the undergraduate dataset and 91.1% for the graduate student’s dataset. The study’s results will aid in implementing targeted interventions and strategies to prevent minority student dropout rates and improve their academic performance.

Digitized world demands data integration systems that combine data repositories from multiple data sources. Vast amounts of clinical and biomedical research data are considered a primary force enabling data-driven research toward advancing health research and for introducing efficiencies in healthcare delivery. Data-driven research may have many goals, including but not limited to improved diagnostics processes, novel biomedical discoveries, epidemiology, and education. However, finding and gaining access to relevant data remains an elusive goal. We identified these challenges and developed an Integrated Radiology Image Search (IRIS) framework that could be a step toward aiding data-driven research. We propose building data bridges to support retrieving ranked relevant documents from integrated repository. My research focuses on biomedical data integration and indexing systems and provide ranked document retrieval from an integrated repository. Though we currently focus on integrating biomedical data sources (for medical professionals), we believe that our proposed framework and methodologies can be used in other domains as well.