Informatics framework for the identification of diagnostic discrepancies and errors

Diagnostic “grey zones” is a term used in pathology, the study of diseases, to describe overlapping morphologic, immunophenotypic and genetic features among various disease subtypes that can lead to diagnostic pitfalls and errors in classifying cancer (e.g. lymphomas). Diagnostic pitfalls are risks that pathologists should be aware of and avoid, and diagnostic errors are failures of medical tests to describe accurately the disease progress in an individual patient. Therefore, pathologists have to perform rigorous medical examinations. These examinations can be used to study diagnostic errors. However, the examinations are documented as unstructured free text. From a computational standpoint, it is…

HOMOLOGY SEQUENCE ANALYSIS USING GPU ACCELERATION

A number of problems in bioinformatics, systems biology and computational biology field require abstracting physical entities to mathematical, computational models. In such studies, the computational paradigms often involve algorithms that can be solved by the Central Processing Unit (CPU). Historically, those algorithms bene- fit from the advancements of computing power in the serial processing capabilities of individual CPU cores. However, the growth has slowed down over recent years, scaling out CPU has shown to be both cost-prohibitive and insecure. To overcome this problem, parallel computing approaches that employ the Graphics Processing Unit (GPU) have gained attention as complementing or replacing…

Using Deep learning method (CNN) for prediction of ubiquitination protein

Ubiquitination, as a post-translational modification, is a crucial biological process presented in cell signaling, death and localization. Identification of ubiquitination protein is of fundamental importance for understanding molecular mechanisms in biological systems and diseases. Although high-throughput experimental studies using mass spectrometry have identified many ubiquitination proteins and ubiquitination sites, the vast majority of ubiquitination proteins remain undiscovered, even in well studied model organisms. To reduce experimental costs, computational (in silico) methods have been introduced to predict ubiquitination sites. If we can predict whether a query protein can be ubiquitinated or not, it is meaningful by itself and helpful for predicting…

Automation of Volumetric Analysis of Adiposity in Canines

Roughly 30-40% of all dogs and cats that are seen by a veterinarian can be classified as obese. Despite this, veterinary practices still utilize a 5 point or 9 point subjective classification system when classifying patients as obese, which can provide difficult when providing accurate nutritional consults to veterinary clients aiming to decrease their pet’s weight. Further, the obesity itself can lead to worsening of comorbid conditions. Thus, an automation of the process of assessing adiposity through CT scan was attempted, looking specifically at the thoracic region of the animal. First, the issues with the current BCS system were highlighted…

Effects of Pain Management Clinical Decision Support in an Inpatient Setting for Patients Experiencing Abdominal Pain

Disorganization of pain-management-related information in an EMR may limit clinicians’ ability to consider clinical factors comprehensively. A clinical decision support (CDS) system for pain management was developed and deployed at University of Missouri Healthcare.  CDS effects were examined for inpatients with diagnoses of diverticulitis, pancreatitis, and abdominal pain.  Statistically significant differences were found in the average NRS-11 self-reported pain scores with a mean reduction of 0.7, and number of pain related medications prescribed, with a mean reduction of 1.2 pain medication orders per day. No statistical correlation was found between the use of the CDS and prescription of different classes of pain medications at discharge, nor…

Use of Powerful Tools for Meaningful Conclusions from Sparse Data

At any given time, over 10 million women are pregnant or lactating in the United States, about 80% of these pregnancies result in a normal pregnancy and life birth. The remaining are associated with a wide range of pregnancy related diseases, an even lower percent of patients present with complications not related to the pregnancy itself. The size of the data is at first glance exciting for the informatics researcher however, the low incidence of positive cases of each type of disease results in sparse data difficult to analyze resulting in less than ideal models for data mining and knowledge…

Using Social Network Analysis and Natural Language Processing to Describe Communication Practices of Interdisciplinary Teams in Primary Care

The Electronic Medical Record (EMR) serves different purposes including documentation of care and billing. One part of the EMR at the University of Missouri Hospital and Clinics is the Message Center. Many people, including healthcare providers, nurses, social workers, therapists, office staff, and nurse care managers (known as the interdisciplinary team, or IDT) work together to deliver healthcare. This research examines how the Message Center is used in primary care by nurse care managers to document care coordination activities, including communication between patients, patient identified family or significant other, and the IDT. Care coordination activities, and the focus of those activities…

Seasonal Influenza Vaccine: Not easy shot to get

During the past nearly 50 years, antigenic variants of subtype H3N2 influenza A viruses have frequently emerged, causing significant public health challenges. The manner in which these variants emerge and their patterns of spread are not well understood. We identified 15 antigenic drift events with 16 antigenic variants during 1968–2016 by using a novel genomic sequence–based antigenicity inference method on ~40,000 H3N2 viruses. New antigenic variants were shown to emerge from certain locations in other continents rather than from Asia alone, and variants emerged year-round and took <2 months to spread across multiple continents. The uncertainty of the location of…

REDESIGN: RDF-based Differential Signaling Framework for Precision Medicine Analytics

Pathway-based analysis holds promise to be instrumental in precision and personalized medicine analytics. However, the majority of pathway-based analysis methods utilize “fixed” or “rigid” data sets that limit their ability to account for complex biological inter-dependencies. Here, we present REDESIGN: RDF-based Differential Signaling Pathway informatics framework. The distinctive feature of the REDESIGN is that it is designed to run on “flexible” ontology-enabled data sets of curated signal transduction pathway maps to uncover high explanatory differential pathway mechanisms on gene-to-gene level. The experiments on two morphoproteomic cases demonstrated REDESIGN’s capability to generate actionable hypotheses in precision/personalized medicine analytics.