LifeHubUK Technical Challenge 2023
NO 1
Intracranial aneurysms are a vascular pathology, that carry a significant risk of morbidity and mortality in the event of rupture.
Computed tomography angiography (CTA) is one of the main imaging modalities for the detection and treatment planning of intracranial aneurysms. The sensitivity rates for detection of aneurysms using CTA ranges between 85–98% compared to digital subtraction angiography (DSA), which is considered as the gold standard for aneurysm imaging. CTA is minimally invasive, widely available, rapid to acquire and has a high resolution, therefore it has the potential to replace DSA.
However small aneurysm detection on CTA remains difficult and misdiagnosis can lead to poor clinical outcome.
Recent advancements in machine learning models have the potential to assist the diagnostic tasks by improving the automated detection and segmentation of intracranial aneurysms. This is of particular interest due to the growing workload of radiologists, which can lead to increased risk of misdiagnosis.
In this challenge we are looking for solutions that can automatically read brain CTA in order to identify and segment intracranial aneurysms.
The challenge requires the submission of a detailed project proposal with the description of a software prototype for the automated identification and segmentation of intracranial aneurysms using CTA.
The description should include an analysis of the algorithm performance with respect to different aneurysm size, location and bleeding severity.
The project proposal should include previous experience, publications in the development of software solutions for the automated detection of intracranial aneurysms.
Bonus: the submission of code developed for the solution of this task using in house data is considered a plus.
NO 2
Aortic Dissection- Calantic Spark Grand Challenge (LifeHub): The aorta is the main artery that carries oxygen-rich blood from the heart to the body. Thus, aortic diseases can pose a significant risk to a patient’s life. Among aortic diseases, aortic dissection is a serious condition in which a tear occurs in the intimal layer of the aorta or bleeding within the aortic wall, resulting in the separation (dissection) of the layers of the aortic wall. Since acute aortic dissection can be fatal, it needs immediate diagnosis and medical attention. An AI tool that can automatically detect and flag the aortic dissection can help the radiologists to prioritize these time sensitive cases. CT angiography (CTA) is the most used modality for diagnosing aortic dissection.
In this challenge, we are looking for solutions that automatically reads chest and abdomen CTAs and look for the signs of aortic dissection to notify the radiologist. Additional information on the condition can be found here Some information can be found here.
Deliverables: Past publications, clinical evidence and any other materials that speak to your expertise on the topic Solution development proposal (approach, timeline, methods) Optional: A proof of concept algorithm trained on data if such is available to you
Criteria for Winner: A jury of technical and commercial experts will carefully evaluate all submitted materials and select one winner based on the depth of field expertise and thoughtfulness of the development proposal.
Contact: lifehubuk@bayer.com
PP-OTH-GB-1022 Feb 2023