HaloScape study on heart attack screening from paper ECG images published on a leading global engineering and technology platform

Health technology company HaloScape has developed a method that can turn a photograph or scan of a paper ECG back into a digital signal and screen it for signs associated with a heart attack. The research has now been published in IEEE Xplore, one of the world's leading digital research libraries for engineering and technology.

The study, “Rule-Based Myocardial Infarction Screening from Paper ECG Images via Digitization and Clinical Criteria,” addresses a simple but persistent problem in healthcare: valuable clinical information often exists only on paper or as an image, with the original digital signal no longer available.

HaloScape's approach uses computer vision to recover ECG waveforms from printed or photographed records and reconstruct them as digital signals that can be analyzed again. The resulting signal is then assessed against predefined clinical criteria associated with myocardial infarction, commonly known as a heart attack.

In practical terms, the method can take an ECG that exists only as a paper record or photograph and turn it back into data that a computer can process.

Rather than relying solely on a black-box AI model to generate an outcome, the study uses clinically defined criteria as part of the assessment process. This makes the signal-generation and screening workflow more transparent and traceable.

The research joins the international scientific literature

The study was first presented at the 34th Signal Processing and Communications Applications Conference, SIU 2026, before being published in IEEE Xplore, bringing the work into the international engineering and technology literature.

IEEE is one of the world's most established and respected professional organizations in electrical engineering, electronics, computer science and technology. Its digital library, IEEE Xplore, is one of the most widely used and prestigious research databases in engineering and academia, providing access to peer-reviewed papers, conference proceedings, technical research and standards.

“Creating meaningful change in healthcare does not always mean generating more data. In many cases, the data we need already exists, but it cannot be used because it is trapped in the wrong format. Decades of valuable health information remain scattered across different systems, devices and, in many cases, paper records. With this study, we focused on making that existing data accessible and meaningful again. Seeing our work published in IEEE Xplore and become part of the international scientific literature is a great source of pride for us. It is also an important reflection of our commitment to putting science and academic rigor at the heart of how we build technology.”

“In health technology, getting the right output is only part of the challenge. We also need to understand what that output is based on and how it was produced. When reconstructing the digital ECG signal from an image, we used deterministic algorithms rather than probabilistic AI models. Our aim was to minimize signal loss during signal generation while avoiding the risk of an AI system estimating or introducing parts of the signal.”

Built around clinical criteria and traceability

The research brings two stages that are often treated separately into a single workflow: recovering a digital ECG signal from an image and assessing that signal using clinical criteria.

First, the photographed or scanned paper ECG is processed using computer vision. The waveform is extracted from the image and reconstructed as a digital signal. The system then evaluates that signal against defined ECG criteria associated with myocardial infarction.

This approach also speaks to a broader challenge in healthcare AI: explainability. The objective is not simply to produce a screening result, but to retain visibility into the clinical indicators and rules that contribute to that result.

For HaloScape, the study is part of a broader effort to make health signals from different sources more accessible, usable and clinically meaningful.

Original paper title: Rule-Based Myocardial Infarction Screening from Paper ECG Images via Digitization and Clinical Criteria
Read the paper: ieeexplore.ieee.org/document/11637005

About HaloScape

HaloScape is a global digital health intelligence platform that brings together health data from everyday life and clinical sources and turns it into meaningful, actionable insights. The platform works across multiple data sources, including wearables, connected health devices, lifestyle and symptom data, biomarkers and clinical health records, helping make health information more understandable and useful for individuals, healthcare professionals and the wider health ecosystem. For more information, visit haloscape.health.

Press contact
press@haloscape.health