Intelliscope

We present a low-cost wireless stethoscope with an artificial intelligence (AI) back-end to help both semi-trained and expert medical professionals to diagnose cardiac abnormalities

Cardiovascular diseases (CVD) are a leading cause of death globally. CVDs include a range of conditions such as coronary artery diseases, heart failures and strokes. WHO reports states that annual death counter due to CVDs are around 17.9 million, and it is around 31% of all deaths world- wide. Early diagnosis of CVDs is crucial for implementing effective treatment plans and prevention of any potential complications. However, criteria like limited health care services, low awareness and excessive costs for diagnostic tests hinder early detection. Thus, the importance of introducing affordable, accessible diagnostic tools is highlighted. Currently, separate phonocardiogram (PCG) and electrocardiogram (ECG) testing devices are used in diagnosis procedures. However, these devices have limitations, such as the fact that one device does not expose all potential heart risks. An integrated ECG-PCG testing device combining both procedures would enhance diagnostic accuracy and sensitivity.

Transfer Learning-Based Approach for Screening Pre-existing Heart Diseases Using Synchronized ECG Signals and Heart Sounds

Diagnosing pre-existing heart diseases early in life is important as it helps prevent complications such as pulmonary hypertension, heart rhythm problems, blood clots, heart failure and sudden cardiac arrest. To identify such diseases, phonocardiogram (PCG) and electrocardiogram (ECG) waveforms convey important information. Therefore, effectively using these two modalities of data has the potential to improve the disease screening process. Here, we evaluate this hypothesis on a subset of the PhysioNet Challenge 2016 Dataset which contains simultaneously acquired PCG and ECG recordings.

Our Study

Our novel Dual-Convolutional Neural Network based approach uses transfer learning to tackle the problem of having limited amounts of simultaneous PCG and ECG data that is publicly available, while having the potential to adapt to larger datasets. Comparisons with methods which used single or dual modality data show that our method can lead to better performance. Furthermore, our results show that individually collected ECG or PCG waveforms are able to provide transferable features which could effectively help to make use of a limited number of synchronized PCG and ECG waveforms and still achieve significant classification performance.

Hardware Device

Intelliscope hardware incorporates dry electrode-based 3-lead ECG integration. This wireless stethoscope has the ability to record synchronized high quality PCG and ECG signals with a powerful AI backend. The ECG add-on is carefully designed with a unique shape, making it very easy to use. The battery powered circuitry inside the control unit, provides filtering and active noise cancellation to the acquired signals and wirelessly transmits them to the user’s mobile phone or computer; allowing the user to record, visualize and analyze the data.