FREEDOM AND SAFETY
Hospitals produce 50 petabytes of data per year. This mass of information comprises clinical notes, lab tests, medical images, sensor readings, genomics, and operational and financial data. At the moment, 97% goes unused – but this is changing, with great potential to transform the quality of medical care.
Here are four ways data analysis is improving healthcare without adding staff or beds.
The X-ray is the oldest form of medical imaging, and still the most commonly used. Chest X-rays alone represent 40% of the 3.6 billion imaging procedures performed worldwide every year. But X-ray “reject rates” – the number of images that cannot be used due to poor image quality or patient positioning – can approach 25%.
To address this, software engineers have developed an application that helps clinicians pinpoint the root causes of rejected images. The app was piloted at the University of Washington Medical Center, and has automated a process that once required 230 mouse clicks and nearly seven hours of work. Reducing reject rates saves time and resources while putting patients on the right path sooner.
In oncology, the process of preparing for, conducting, and documenting tumor board meetings is frequently suboptimal and non-standardized. Each specialist aggregates data on a patient in a silo. As a result, meetings are spent switching back and forth between the different systems and technologies used across each discipline.
To address this, an alliance between Roche Diagnostics and GE Healthcare is combining and analyzing patients’ diagnostic data - including genomics, tissue pathology, and biomarkers - with their medical imaging and monitoring data. From here, cloud-based data integrating software could fundamentally change the process of tumor board meetings, helping doctors make more informed, faster diagnoses and individualize treatments to each patient.
Integrating data can have transformative effects across the entire healthcare ecosystem. GE Healthcare recently began a partnership with Vanderbilt University Medical Center (VUMC), to enable safer, more-precise immunotherapies.
The project will retrospectively analyze and correlate the immunotherapy responses of thousands of cancer patients with their demographic, genomic, tumor, cellular, proteomic, and imaging data. From here, AI-powered apps will help physicians identify the most suitable treatment for each patient.
A small but growing number of hospitals are implementing NASA-style mission control Command Centers to manage their functions and services. The goal is to address the capacity, safety, quality, and wait-time issues that have plagued healthcare.
A hospital Command Center pulls in streams of data from various systems, generating analytics that help staff predict what will happen in the next 24 to 48 hours. The data is displayed on Command Center screens and on tablets and mobile devices. This allows staff to focus on delivering care, rather than organizing it.