We're helping forward thinking partners revolutionize healthcare.
From volume forecasting in the Emergency Department to planning for hospital bed capacity. Explore our most recent projects to see how your team can benefit from working with LKS-CHART.
Automated Analysis of Cervical Spine CT Scans
Developing A 3D convolutional neural network (CNN) that will classify CT scan images of the cervical spine as containing or not containing a fracture.
Tuberculosis (TB) Clinic Database
Working with the Tuberculosis (TB) clinic to implement Natural Language Processing that extracts patient information from dictated notes, making it readily available for analysis.
General Internal Medicine Resident Staffing Redesign
Building a simulation model based on the current workload of the internal medicine residents, to re-assign residents to match patient volumes.
Emergency Department Analytics
Creating analytical models for the Emergency Department (ED), including volume forecasting, alerting when full capacity will be reached, and optimizing patient flow.
CHART-IPBR Fellowship Optimizing Antimicrobials
Working with the Interprofessional Practice Based Research (IPBR) team to create an automated tool to identify patients who are eligible for oral antibiotics.
Improving inpatient rates of oral anticoagulation for stroke prevention in Atrial Fibrillation (IMPROVE-AF)
Explore the feasibility and initial performance of an algorithm for the detection of inpatients with atrial fibrillation or flutter who are candidates for oral anticoagulation.
ED Nurse Location Assignment Tool
Building a mathematical model that produces a repetition-minimizing assignment, and designing an interactive web application to facilitate the daily assignment process.
A Vascular Surgery Quality Improvement Initiative
Tracking patient outcome measures in vascular surgery, including in-hospital stroke, discharge medications, and follow-up encounters, so surgeons can identify areas for improvement.
An Early Warning System for General Internal Medicine
Using near-real-time hospital data to help clinicians identify high-risk patients so they can improve patient care and reduce the chances of mortality.
Cath Lab Recovery Bed Capacity Planning
Using a discrete event simulation model, it will analyze historical volumes of patients needing recovery beds, as well as, predict volume growth, to project the number of needed recovery beds for now and the next 10 years.
Length of Stay (LOS) Predictions using Text Notes
Using available inpatient admission notes, a Latent Dirichlet Allocation (LDA) model was used to extract features from raw text data to predict patient length of stay.