Zdravíme

Problem: Identifying high-risk patients for CKD screening poses a critical challenge. Our analysis of hospital data reveals a significant gap in proactive care, leaving potentially life-threatening kidney diseases undetected until late stages. For that we only have an observational dataset from the past patient records which is not samped at random or representative, because the albumin screening is adopted at different rates in different medical specializations.

Solution: We developed a system that analyses patient data to prompt doctors, ensuring timely albuminuria screenings for at-risk individuals. While doing this we also analysed local IKEM data for the prevalence of CDK-at-risk patients.

Impact: By targeting high-risk patients efficiently, our system aims to detect kidney disease earlier, empowering timely interventions, and potentially saving lives. By implementing this system universally, we can help prevent life-threatening conditions in chronic patients. We can offer every individual the opportunity for a fuller, healthier life through early detection and intervention.

Feasibility & financials: Our scalable solution would integrate seamlessly within existing hospital infrastructure as we used already existing patient data, ensuring a cost-effective implementation across healthcare facilities. The investment in preventive care today holds the promise of significant cost savings and, more importantly, improved patient outcomes tomorrow.

What is new about your solution?: Our solution's novelty lies in its potential universality. Imagine the collective impact if every hospital adopts this proactive approach. It's not just about data analysis; it's about equipping healthcare providers with a powerful tool to safeguard lives on a grand scale.

What you have built at the hackathon - text explanation + code (e.g. GitHub link): In the course of hackathon we had to process complicated data, communicate the problems and their importance with the IKEM medical staff, perform our analyses and prepare a presentation. There are two specific outcomes of our work: 1) a model, that can predict the risk of high ACR from other diagnoses and patient data at IKEM, 2) a prevalence analysis of patients at IKEM. https://github.com/Jan-Blaha/HHH23-CKD

What you had before the hackathon, please mention open source as well: We did not have any prior experience with the data or domain. We did not use any complex analytical tools beyond basic open-source data/statistics libraries. We did rely heavily on statistical skills in handling problems, which look simple but lead to trouble not handled properly.

What comes next and what you wish to achieve: Health! Helping people!

Video: https://www.youtube.com/embed/jqUI347GzeQ