Predicting Pulmonary Hypertension by Electrocardiograms Using Machine Learning

Pulmonary hypertension (PH) is a condition of high blood pressure that affects the arteries in the lungs and the right side of the heart (Mayo Clinic, 2017). A mean pulmonary artery pressure greater than 25 mmHg is defined as Pulmonary hypertension. The estimated 5-year survival rate from the time of diagnosis of pulmonary hypertension is only 57% without therapy and patients with right heart failure only survive for approximately 1 year without treatment (Benza et al., 2012). Given the indolent nature of the disease, early detection of PH remains a challenge leading to delays in therapy. Echocardiography is currently used as a screening tool for diagnosing PH. However, electrocardiography (ECG), a more accessible, simple to use, and cost-effective tool compared to echocardiography, is less studied and explored for screening at-risk patients for PH. The goal of this project is to create a neural network model which can process an ECG signal and detect the presence of PH with a confidence probability. I created a dense neural network (DNN) model that has an accuracy of 98% over the available training sample. For future steps, the current model will be updated with a model suited for time-series data. To balance the dataset with proper training samples, I will generate additional data using data augmentation techniques. Through early and accurate detection of conditions such as PH, we widen the spectrum of innovation in detecting chronic life-threatening health conditions and reduce associated mortality and morbidity.

In Silico Prediction of Blood-Brain Barrier Permeability of Chemical Compounds through Molecular Feature Modeling

The introduction of computational techniques to analyze chemical data has given rise to the analytical study of biological systems, known as "bioinformatics". One facet of bioinformatics is using machine learning (ML) technology to detect multivariable trends in various cases. Amongst the most pressing cases is predicting blood-brain barrier (BBB) permeability. The development of new drugs to treat central nervous system disorders presents unique challenges due to poor penetration efficacy across the blood-brain barrier. In this research, we aim to mitigate this problem through an ML model that analyzes chemical features. To do so: (i) An overview into the relevant biological systems and processes as well as the use case is given. (ii) Second, an in-depth literature review of existing computational techniques for detecting BBB permeability is undertaken. From there, an aspect unexplored across current techniques is identified and a solution is proposed. (iii) Lastly, a two-part in silico model to quantify likelihood of permeability of drugs with defined features across the BBB through passive diffusion is developed, tested, and reflected on. Testing and validation with the dataset determined the predictive logBB model's mean squared error to be around 0.112 units and the neuroinflammation model's mean squared error to be approximately 0.3 units, outperforming all relevant studies found.