Other exhange programmes

AI Fellowship Programme for students at the Lithuanian University of Health Sciences

Are you passionate about advancing your knowledge in medical research and innovation? As part of the SustAInLivWork project, we are proud to offer an international AI Fellowships within our research group at the Automation of Cardiovascular Investigation Laboratory at the Institute of Cardiology, Lithuanian University of Health Sciences (LSMU). This programme aims to foster innovation and research excellence by focusing on the transformative potential of artificial intelligence in medical and healthcare applications. Below is an overview of potential research topics related to “AI for Healthcare”. For detailed application guidelines, please visit the  homepage.

Fellowship Highlights

As a fellow in our programme, you will engage in hands-on research projects in collaboration with multidisciplinary teams, including clinicians. Our programme offers:

• Access to cutting-edge computational resources, including medical images.
• Opportunities to contribute to and expand our existing projects.
• Support in developing and testing innovative AI solutions in real-world clinical settings.

Fellows are encouraged to bring their own ideas to shape impactful research in domains such as automated diagnostics, and predictive healthcare analytics.

About Us

As the largest institution of higher education for biomedical sciences in Lithuania, LSMU integrates education, research, and healthcare delivery. Our laboratory is at the forefront of cutting-edge research in cardiovascular health, focusing on the development and application of automated and digital technologies for diagnosing and managing cardiovascular diseases. The majority of our research is conducted in close collaboration with clinicians, ensuring a direct connection to patient care and practical application of our findings. As part of the Institute of Cardiology at LSMU, we collaborate on various national and international research projects funded by the Research Council of Lithuania and other prestigious organisations.

INTERSHIP FOCUS AREAS

Participants will have the opportunity to work on:

Application of Deep Neural Networks for Automated Echocardiographic Heart Failure Diagnosis

Contact person: Arnas Karužas, MD arnas.karuzas@lsmu.lt

Summary: Heart failure is a global health challenge, with early detection being critical for effective treatment and improved patient outcomes. Echocardiography is a widely used, non-invasive imaging modality for assessing cardiac function. However, interpreting echocardiographic data is often complex and requires specialised expertise, leading to potential variability in diagnoses.

This project focuses on employing deep neural networks (DNNs) to automate the detection and classification of heart failure from echocardiographic images. By training AI models on large datasets of annotated ultrasound images, the system aims to identify key indicators of heart failure, such as reduced ejection fraction or abnormal wall motion, with high accuracy and consistency.

Challenges include the variability in image quality due to operator dependence, differences in ultrasound machines, and patient-specific factors. The project addresses these issues by developing robust preprocessing techniques and leveraging advanced architectures such as convolutional neural networks (CNNs) and recurrent neural networks (RNNs).

Artificial Intelligence for Cardiac CT Scan Stenosis Detection

Contact person: Dovydas Verikas, MD dovydasverikas@lsmu.lt

Summary: Cardiovascular diseases are among the leading causes of morbidity and mortality worldwide, making early and accurate diagnosis crucial. Cardiac CT scans are a valuable tool for detecting coronary artery stenoses; however, manual analysis is time-consuming and prone to interobserver variability. Our research focuses on leveraging artificial intelligence to automate the detection and quantification of stenoses in cardiac CT images.

By integrating advanced image processing algorithms and deep learning techniques, the project aims to improve diagnostic accuracy and efficiency. AI models are trained to identify and segment coronary arteries, detect stenotic regions, and quantify their severity. This automation reduces the burden on clinicians, enabling faster diagnoses and better patient outcomes.

Challenges addressed include variability in image quality, artifacts from patient movement, and the need for robust generalisation across diverse datasets. The project also explores explainable AI methods to ensure transparency and trust in clinical decision-making.

Topic: Unsupervised Learning for Stroke and Myocardial Infarction Data Analysis

Contact person: Gintarė Šakalytė, MD PhD gintare.sakalyte@lsmu.lt

Summary: Stroke and myocardial infarction are the leading causes of death and disability worldwide, requiring timely and accurate analysis of patient data for effective prevention and management. Traditional approaches often rely on supervised learning methods that require extensive labelled datasets, which can be resource-intensive to prepare.

This project explores the potential of unsupervised learning techniques to uncover hidden patterns and relationships in stroke and myocardial infarction datasets. By utilising clustering, dimensionality reduction, and anomaly detection algorithms, the research aims to identify patient subgroups, novel biomarkers, and risk factors without prior labelling.

Key challenges include handling heterogeneous data types, such as clinical records, imaging, and biochemical markers, as well as ensuring the interpretability of the findings. Advanced methodologies like autoencoders, self-organising maps, and contrastive learning are applied to address these challenges.

Programme Details:

• Duration: The internship should last for at least one calendar month.
• Location: Automation of Cardiovascular Investigation Laboratory, Institute of Cardiology, LSMU, Kaunas, Lithuania.
• Start Date: The screening of submitted applications is performed 4 times a year: at the beginning of September, December, March, and June.

Eligibility requirements

• Student in Bachelor (from 3rd semester) or Master study cycle.
• Have AI background, solid programming and IT skills.
• Have a minimum grade point average of 8+.
• Have an English language level B2 or above.

Interested candidates are encouraged to check on the SustAInLivWork homepage where you will find all relevant information on the application process. We look forward to welcoming you to our team.

If you have any questions about the programme or how to apply, please contact us at dovydas.verikas@lsmu.lt.

LiMSA is a non-governmental, non-profit organization that unites biomedical students from Vilnius and Kaunas. LiMSA is a unique platform for international exchange opportunities with more than 50 countires in 6 world‘s regions. 

More about LiMSA membership and international exchange opportunities: https://limsa.lt/ (In Lithuanian).