PLOS Digital Health, a new journal driving transformation in the delivery of equitable and unbiased healthcare

Sepsis may be misdiagnosed if a patient is not thoroughly assessed or the symptoms are misinterpreted, which can lead to serious health complications or even death. To improve understanding of disease progression, this work identified distinct states in sepsis in terms of their extremal clinical manifestations, also known as archetypes. Six states were identified, each characterized by a unique set of pathological responses that can be mapped back to organ function(s), along with an association between patient attributes and sepsis states.

This paper describes the development and implementation of the Hyperion software platform to address five data management challenges of an academic cancer center: lowering the skill level required to maintain the system, reducing costs, allowing users to access data autonomously, optimizing data security and utilization, and shifting technological team structure to encourage rapid innovation.

The research team developed a 2D convolutional neural network for accurate scar quantification of cardiac magnetic resonance (CMR) late gadolinium enhancement (LGE) images. This methodology showed reliability and high potential for clinical adaptability.