Researchers from Terra Quantum, a leading quantum technology company, have developed a new image analysis system that outperforms current methods in identifying healthy livers for transplantation while protecting patient privacy.
The hybrid quantum neural network model combines classical machine learning systems with quantum computing techniques to identify moderate and severe non-alcoholic fatty liver disease with fewer false positives and greater accuracy than imaging experts and traditional algorithms. This breakthrough has the potential to increase the overall transplant rate by making it easier to identify donors with healthy livers.
The research, published in the peer-reviewed journal Diagnostics, was led by medical experts at the University of Trieste and involved collaboration with Terra Quantum’s quantum system engineer Luca Lusnig. Markus Pflitsch, founder and CEO of Terra Quantum, highlighted the value of combining classical and quantum software to achieve accurate results. Dr. Fabio Cavalli, co-author of the article, emphasized that this approach provides sufficient data for analysis without compromising patient privacy.
Quantum Algorithm Outperforms Classical Methods in Identifying Healthy Livers for Transplant
A recent study published in the peer-reviewed journal Diagnostics has demonstrated a novel image classification system that leverages quantum computing techniques to identify healthy livers suitable for transplantation with higher accuracy than classical methods. This hybrid quantum neural network (HQNN) model, developed by Terra Quantum in collaboration with medical and informatics experts, combines the strengths of classical and quantum software to provide a more accurate diagnosis.
The HQNN model uses real-world clinical data to identify moderate and severe non-alcoholic fatty liver disease (NAFLD) with fewer false positives and greater accuracy than imaging experts and traditional algorithms. NAFLD is a leading cause of liver disease worldwide, affecting an estimated 32% of adults, and fatty liver disease steatosis is often the single most important variable determining the success of a liver transplantation.
The research team’s goal was to analyze images and categorize organs into two groups: livers healthy enough for transplant and unhealthy livers likely to fail if transplanted. The HQNN model achieved an image classification accuracy of 97%, surpassing traditional methods by 1.8%. Moreover, the hybrid model outperformed human experts when sharing data from multiple hospitals, achieving over 90% accuracy with a false negative rate below 5%.
Protecting Patient Privacy while Sharing Data
One of the significant advantages of this HQNN model is its ability to comply with privacy laws without compromising accuracy. The research simulated a real-world scenario in which multiple hospitals wanted to create a common model based on data from each individual hospital without physically sharing the data with each other or an external server. This approach, known as federated learning, enables collaborative training across multiple clients without exposing sensitive data.
“This new approach will provide sufficient data for analysis with significant results without having to transfer large quantities of data and at the same time protecting patient privacy,” said Dr. Fabio Cavalli, co-author of the article and founder of the Paleoradiology and Related Sciences Research Unit of Tripei. This advance has the potential to make it easier for hospitals to share data while complying with existing and pending regulations, such as the EU AI Law.
Potential Impact on Liver Transplantation
The HQNN model’s ability to identify healthy livers with higher accuracy than classical methods could increase the overall transplant rate. Living donors can donate part of their liver to an individual who needs a transplant, and the donor’s liver will grow back to its original size within a few months. By making it easier to identify donors with healthy livers, this model could lead to more successful transplants.
Moreover, this approach enables hospitals to share data and still comply with privacy regulations, which is particularly important for international multi-center works. “This research allows us to participate in international multi-center works, greatly shortening the bureaucratic and technical-organizational obligations imposed by the GDPR to guarantee the quality of the data and the respect for data protection rights,” said Graziano De’ Petris, co-author of the article and data protection manager of three healthcare entities in the region.
Future Applications of Hybrid Quantum Neural Networks
The results of this study suggest that HQNNs, combined with a federated learning framework, could become the preferred model over centralized systems for future applications where data privacy and availability are of concern. Due to their ability to capture intricate correlations between data, this architecture could become the standard technique for creating computer-aided diagnosis systems.
Terra Quantum, a leading quantum technology company based in Germany and Switzerland, is at the forefront of developing innovative solutions that leverage the power of quantum computing. Their “Quantum as a Service (QaaS)” platform provides customers with access to an extensive library of algorithms, including hybrid quantum optimization and hybrid quantum neural networks, which can be used for solving complex logistics problems or pattern recognition, among other things.
External Link: Click Here For More
