Understanding heart function and illness, as well as testing new medications for heart problems, has traditionally been a difficult and time-consuming endeavour. Using cellular and engineered tissue models in a dish is a potential technique to investigate disease and test new medications. However, conventional methods to examine cardiac cell contraction and calcium handling are time-consuming, error-prone, and require expensive specialist equipment.

There is a clear medical need for a more efficient, accurate, and accessible approach to studying heart function that employs AI and machine learning. Columbia Engineering researchers have introduced a novel tool that directly tackles these difficulties. 

BeatProfiler

BeatProfiler is a new tool for rapidly analyzing heart cell function. It automates the analysis of heart cell function using video data. It is the first system to combine the study of various heart function indicators, including contractility, calcium handling, and force output, into a single tool. This integration dramatically accelerates the analysis process and minimizes the risk of errors.

BeatProfiler helped researchers differentiate between various disorders and their respective levels of severity. Additionally, it allowed for the quick and unbiased evaluation of medications that impact heart function. Similar to a significant portion of Vunjak-Novakovic's research, this initiative was motivated by a clinical imperative to expedite and enhance the diagnosis of cardiac disorders. This project underwent a lengthy development process spanning several years, during which the team gradually incorporated various features individually. 

Cardiac models

The primary objective was to create a tool that could effectively measure the function of the cardiac models used by the team to study heart diseases and evaluate potential treatments. However, the researchers also had an immediate requirement to promptly and precisely assess the function of their cardiac models in real time.

The lab's advancements, including the milliPillar and multiorgan tissue models, led to the production of more cardiac tissues. However, the improved capabilities of these tissues necessitated the development of a faster method to measure the function of cardiomyocytes and tissues. It was done to facilitate research on genetic cardiomyopathies, cosmic radiation, immune-mediated inflammation, and drug discovery.

Evaluation

The study demonstrated that BeatProfiler exhibited precise analysis of cardiomyocyte function, surpassing current techniques in speed—achieving up to 50 times faster results in certain instances—and reliability. It identified nuanced alterations in the mechanical response of artificially created heat tissue that alternative instruments could overlook.

Conclusion

The research team is currently focused on enhancing BeatProfiler's functionalities to cater to many applications in cardiovascular research, encompassing a wide range of heart illnesses that impact cardiac pumping and drug discovery. BeatProfiler is being tested and verified on numerous in vitro cardiac models, including synthetic heart tissue models, to ensure its relevance to various research questions.

In addition, they are enhancing their machine-learning system to broaden its application to various cardiac conditions and drug impact categorization. The ultimate objective is to modify BeatProfiler for implementation in pharmaceutical environments and expedite the evaluation of numerous potential medications simultaneously.

Researchers information

• Youngbin Kim, Department of Biomedical Engineering, Columbia University, New York, NY, USA
• Kunlun Wang, Department of Biomedical Engineering, Columbia University, New York, NY, USA
• Roberta I. Lock, Department of Biomedical Engineering, Columbia University, New York, NY, USA
• Trevor R. Nash, Department of Biomedical Engineering, Columbia University, New York, NY, USA
• Sharon Fleischer, Department of Biomedical Engineering, Columbia University, New York, NY, USA
• Bryan Z. Wang, Department of Biomedical Engineering, Columbia University, New York, NY, USA
• Barry M. Fine, Department of Medicine, Division of Cardiology, Columbia University Medical Center, New York, NY, USA
• Gordana Vunjak-Novakovic, Department of Biomedical Engineering, Columbia University, New York, NY, USA, Department of Medicine, Division of Cardiology, Columbia University Medical Center, New York, NY, USA

Source: https://ieeexplore.ieee.org/document/10490213

Image source: Unsplash