Mälardalen University, Sweden
Title: Potentials of the intelligent phonocardiography as an emerging approach in cardiac assessments
Biography: Arash Gharehbaghi
Recent progresses in artificial intelligence made development of efficient decision support systems feasible. Application of such the DSS is rather seen in primary healthcare centers where accuracy of cardiac diagnosis is substantially low because of the complexities of cardiac auscultation. Our longstanding studies on heart sound analysis resulted in the novel methods that can provide sufficient means to extract significant medical information from the sounds to help the physicians in decision making. These methods were incorporated into a stand-alone system composed of an electronic stethoscope in conjunction with a portable computer. The resulted system, which we called the Intelligent phonocardiography (IPCG), provides an easy-to-use and inexpensive approach for cardiac assessments. Both the accuracy and the sensitivity of the IPCG in screening children with congenital heart disease were estimated to be higher than 87.0%, when a patient population of more than 250 individuals was employed. In a separate study, performance of the IPCG was investigated for assessing severity of valvular aortic stenosis in elderly patients, and the reliability and accuracy of the approach were estimated to be more than 80%. It is worth noting that screening patients with aortic stenosis based on IPCG had already been studied, where an accuracy of higher than 85% was achieved. Potential of the IPCG for pediatric cardiac assessments was rather studied in disease identification and also in discrimination between different cardiac defects with the systolic murmurs. Screening of the children with isolated bicuspid aortic valve, ventricular septal defect, and discrimination between valuvar aortic and pulmonic stenosis are considered as the examples of such studies. Results show that the IPCG has a high potential to be used in primary healthcare centers as an efficient decision support system. This can drastically reduce unnecessary echocardiography which is by far a more expensive approach.
- Gharehbaghi A, et al (2015) Assessment of aortic valve stenosis severity using intelligent phonocardiography. International Journal of Cardiology 198:58-60.
- Gharehbaghi A, et al (2017) A Deep Machine Learning Method for Classifying Cyclic Time Series of Biological Signals Using Time-Growing Neural Network, under press.
- Sepehri A, et al (2016) An intelligent phonocardiography for automated screening of pediatric heart diseases, Journal of Medical SYstems 40(1).
- Gharehbaghi A, et al (2017), Intelligent phonocardiography for screening ventricular septal defect using time growing neural network, Informatics Empowers Healthcare Transformation 238:108.
- Gharehbaghi A, et al (2015) An intelligent method for discrimination between aortic and pulmonary stenosis using phonocardiogram, World Congress on Medical Physics and Biomedical Engineering 1010-1013.
- Gharehbaghi A, et al (2015) A novel method for screening children with isolated bicuspid aortic valve, Cardiovascular Engineering and Technology, 6(4):546-556.