EUROPEAN HEALTHCARE HACKATHON
19 - 21 NOV 2021 IKEM
European Healthcare Hackathon 2021
EHH is an international hackathon tackling some of the most pressing healthcare challenges today. We bring hackers, students, entrepreneurs, healthcare, and industry experts together to create new ideas and technologies for diabetology, surgery, transplantology, patient care and comfort. You can apply for free, as an individual, or as a team of 3 members.
JULY – We are ON. Pre-register and be the first one to know. ✔️
OCTOBER – Registrations OPEN – Form a team of 3 members or apply as an individual and find a team on the spot. ✔️
NOVEMBER – New deadline for registration is 15 NOV 2021 – Only a couple of places for the OFFLINE track available.
NOVEMBER – Invitations SENT – Watch out for an invite email with further instructions.
19 – 21. NOVEMBER 2021
Online track – You or your team will work on challenges remotely without being physically present at the hackathon venue in IKEM.
Offline track – We will be inviting up to 100 hackers (including teams and individuals) that will work on the challenges directly at IKEM in Prague. We will meet at Institut klinické a experimentální medicíny (Institute of Clinical and Experimental Medicine) – IKEM in Prague together with practitioners and other healthcare and technology experts who will be your mentors during the 48 hours hackathon.
18 NOV – Thursday – We will meet online with some partners of the hackathon. More info coming.
HACKING DAY 1
19 NOV – Friday – The hackathon starts at 9:00 AM with the opening ceremony for both online and offline hackers. Hacking will start at 1:00 PM.
HACKING DAY 2
20 NOV – Saturday – Hacking continues. You will have a chance to join workshops and seminars organized by our partners.
21 NOV – Sunday – Before 12:00 PM both online and offline hackers will submit their hacks and judges will start with the evaluation. At 4:30 PM the winner announcement ceremony will start.
1st PLACE - 3 000 EUR
2nd PLACE - 2 000 EUR
3rd PLACE - 1 000 EUR
Edwards Lifesciences Prize - 2 000 EUR
InterSystem Prize - 1st place 1 500 EUR, 2nd place 1 000 EUR and 3rd 500 EUR
AstraZeneca Prize - Grant of 7 800 EUR for a six-month collaboration
and more partner prizes to be announced
- MORE INFO COMING SOON -
1. Heart rate detection using mobile phone
Detect the heart pulse from the finger on the hand, detect any irregularity in possible fibrillation if the device learns C9the. The purpose is to help simple self-diagnosis.
2. Measurement of urinary continence in hospitalized patients
Hospitalized patients should have their urine output monitored. The aim is to develop a solution to measure continuous urine output. This is a measurement in patients who have a urinary catheter in place. It is possible to consider a solution that will weigh the urine volume or some form of a flow meter.
3. Mobile navigation around IKEM (also for the blind)
Due to the completion of IKEM, the possibilities of moving around the IKEM campus are constantly changing. Therefore, we would like to create building navigation that will suggest the optimal route based on the position of the person (even if he/she gets lost) and the specified place where he/she is heading. An important aspect is an ease with which the client can change the layout (closures, impassabilities, disabled lifts, etc.). The application should be user-friendly for the blind in the second step.
4. National Register/ IVLP prescriptions
Creation of a national prescription register system for IVLP resources, Web application available for doctors with the possibility of searching prescriptions, prescription management for pharmacists …
5. Volumetry for TX of the liver in VR
IKEM uses its own solution for the visualization of complex surgical procedures on the liver in the Virtual Reality environment. The goal is to devise an algorithm that can compute the volume of surface polygonal non-symmetric 3D models (tetrahedron) in the Unity 3D environment. Part of this challenge may include a cutting surface that will be able to split two objects, recalculate their facets, vertexes, normals, and create two separate mesh objects / gameObjects.
6. Wound image processing /Thermal imaging camera image comparison
In the early diagnosis of complications in diabetic patients, it is necessary to compare the images of both feet of the lower limbs and evaluate any variations in temperature that indicate potential problems. Evaluation of the degree of defect of granulation and necrosis. Should be also able to compare 2 or more consecutive images and evaluation of healing.
7. zCase – forecasting probalibility of ambulace arrival
Based on data from zCase + information about weather, location, time of day, date, year season it should calculate the probability of the hospital’s emergency admission load.
8. AI in Health
To be able to predict what will happen in the future, clinicians need to analyze and combine a huge variety of data, both structured and unstructured. Can you give them a hand?
9. CKD - the stealth assassin whose symptoms don't hurt
CKD is currently a much-neglected topic, but it affects 11% of the population (about 1.2M people in the Czech Republic), where 6.000 patients a year go straight from the street to dialysis. The biggest risk is that many patients don’t even know they have the disease and when they do find out, it may be too late for treatment. Let’s make sure that the disease is found out early so that effective treatment can be started and the patient’s quality of life with functioning kidneys can be maintained. Let’s explore avenues from both patients who don’t know much about the disease and from physicians (diabetologists, nephrologists, internists, and general practitioners) that we can help alert to patients at risk.
AstraZeneca is ready to support the project financially even after the hackathon in the form of an AstraZeneca prize „grant“ of 200.000 CZK for a six-month collaboration for the team that points to a realistic path.
10. Aortic Stenosis Detection from electrocardiogram data
Cardiovascular diseases are the leading cause of death for both men and women all over the world and we, together with your help would like to change that.
From available data we see, that the majority of people with AS (aortic stenosis) do not receive treatment simply because they are never diagnosed with the disease. For a person to be diagnosed he/she needs to have an echocardiogram which is not a very common examination and is not easily available to all patients. We, with your help, would like to try and develop an algorithm that uses electrocardiograms (a widespread examination that the majority of cardiovascular patients receive) to detect severe aortic stenosis. Are you ready to save lives? The time is now!
11. Aortic Stenosis Detection via cNLP (clinical natural language processing)
Over the last decades, clinical record keeping changed significantly, switching from paper to electronic health records (EHRs). And while some of those are codified in structured fields of EHRs, the great majority of relevant clinical information appears embedded within unstructured narrative free-text.
With your help, we would like to develop reliable cNLP (clinical natural language processing) that would scan those free-text evaluations done by physicians.
Even when a patient is diagnosed with AS (aortic stenosis), he/she may not receive treatment because he/she may not be referred for treatment. Utilizing the cNLP we would like to help those patients receive the treatment they need.
12. Aortic Stenosis Detection from non-invasive blood pressure waveforms
The goal of this challenge is to detect severe aortic stenosis using continuous non-invasive blood pressure waveforms. The given dataset includes pressure-derived features, demographics, and medical history from patients (some with and some without aortic stenosis). In this challenge, you will use this data to train and validate your model for detecting patients with severe aortic stenosis. After finalizing your model, we will test the performance of your model on our private, held-out data set.
14. 75 and thriving!
Remember when that Nugget Tweet touched the hearts of 3.2 million of us? Your genius could potentially touch over 2 million hearts; quite literally.
We call for a reliable algorithm and tool to connect elderly TAVI patients remotely with their TAVI hospitals, to alert them of specific symptoms of potential health risk after discharge. Your genius, our TAVI innovations, and a pioneering optimized TAVI program will enable a safe and fast transition to post-TAVI life. Somewhere a granddad is uncertain about attending the graduation of their grandchildren, let’s make their day, shall we?
15. Heart Failure Prediction
To more effectively treat patients and reduce the burden on the healthcare system, we propose a machine learning model to identify patients who are likely to have a bad outcome (ICU admission, 30-day readmission, 30-day rehospitalization, or death within 30 days) so that clinicians are alerted to the severity of a patient’s case and can better allocate resources and treat patients. Data from 350 patients enrolled in the study to collect data for this project will be available.