A Novel Artificial Intelligence Driven Adaptive Acoustic Profiling Technique for Noninvasive Cardiovascular Diagnostics

R. Srinivasan; K. S. Guruprakash; L. Manikandan; R. Narendiran; Mufti Mahmud

doi:10.1201/9781003541011-9

A Novel Artificial Intelligence Driven Adaptive Acoustic Profiling Technique for Noninvasive Cardiovascular Diagnostics

R. Srinivasan^*
, K. S. Guruprakash
, L. Manikandan
, R. Narendiran
, Mufti Mahmud

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Chapter › peer-review

Abstract

Artificial Intelligence (AI) has revolutionized medical diagnostics, notably enabling noninvasive cardiovascular health evaluations. This study introduces Adaptive Acoustic Profiling (AAP), an AI-based approach using a hybrid Convolutional Recurrent Neural Network (CRNN) to detect micro-acoustic blood flow anomalies. Leveraging the public PhysioNet MIMIC-III database, AAP achieved 94.2% diagnostic sensitivity (CI: 92.8–95.6%), outperforming conventional Doppler ultrasound at 89.1% (CI: 87.0–91.2%). It reduced false-positive rates by 23% without compromising sensitivity (96.8%), demonstrated robust cross-demographic performance, and introduced the Hemodynamic Complexity Index (HCI), showing an 87% correlation with early vascular disorders. With 38% faster computational efficiency, delivering results in under 15 seconds, AAP offers transformative potential in precision medicine and early cardiovascular care.

Original language	English
Title of host publication	Computer Vision in Healthcare Prediction, Detection and Diagnosis
Publisher	CRC Press
Pages	127-140
Number of pages	14
ISBN (Electronic)	9781040525456
ISBN (Print)	9781032864815
DOIs	https://doi.org/10.1201/9781003541011-9
State	Published - 1 Jan 2026
Externally published	Yes

Bibliographical note

Publisher Copyright:
© 2026 Saurav Mallik, Sandeep Kumar Mathivanan, Prabhu Jayagopal, Hong Qin and Ben Othman Soufiene.

ASJC Scopus subject areas

General Social Sciences
General Biochemistry, Genetics and Molecular Biology
General Engineering
General Mathematics
General Agricultural and Biological Sciences
General Pharmacology, Toxicology and Pharmaceutics

Access to Document

10.1201/9781003541011-9

Cite this

@inbook{e42f4d1409f547d083d88641651409d6,

title = "A Novel Artificial Intelligence Driven Adaptive Acoustic Profiling Technique for Noninvasive Cardiovascular Diagnostics",

abstract = "Artificial Intelligence (AI) has revolutionized medical diagnostics, notably enabling noninvasive cardiovascular health evaluations. This study introduces Adaptive Acoustic Profiling (AAP), an AI-based approach using a hybrid Convolutional Recurrent Neural Network (CRNN) to detect micro-acoustic blood flow anomalies. Leveraging the public PhysioNet MIMIC-III database, AAP achieved 94.2\% diagnostic sensitivity (CI: 92.8–95.6\%), outperforming conventional Doppler ultrasound at 89.1\% (CI: 87.0–91.2\%). It reduced false-positive rates by 23\% without compromising sensitivity (96.8\%), demonstrated robust cross-demographic performance, and introduced the Hemodynamic Complexity Index (HCI), showing an 87\% correlation with early vascular disorders. With 38\% faster computational efficiency, delivering results in under 15 seconds, AAP offers transformative potential in precision medicine and early cardiovascular care.",

author = "R. Srinivasan and Guruprakash, \{K. S.\} and L. Manikandan and R. Narendiran and Mufti Mahmud",

note = "Publisher Copyright: {\textcopyright} 2026 Saurav Mallik, Sandeep Kumar Mathivanan, Prabhu Jayagopal, Hong Qin and Ben Othman Soufiene.",

year = "2026",

month = jan,

day = "1",

doi = "10.1201/9781003541011-9",

language = "English",

isbn = "9781032864815",

pages = "127--140",

booktitle = "Computer Vision in Healthcare Prediction, Detection and Diagnosis",

publisher = "CRC Press",

}

TY - CHAP

T1 - A Novel Artificial Intelligence Driven Adaptive Acoustic Profiling Technique for Noninvasive Cardiovascular Diagnostics

AU - Srinivasan, R.

AU - Guruprakash, K. S.

AU - Manikandan, L.

AU - Narendiran, R.

AU - Mahmud, Mufti

PY - 2026/1/1

Y1 - 2026/1/1

N2 - Artificial Intelligence (AI) has revolutionized medical diagnostics, notably enabling noninvasive cardiovascular health evaluations. This study introduces Adaptive Acoustic Profiling (AAP), an AI-based approach using a hybrid Convolutional Recurrent Neural Network (CRNN) to detect micro-acoustic blood flow anomalies. Leveraging the public PhysioNet MIMIC-III database, AAP achieved 94.2% diagnostic sensitivity (CI: 92.8–95.6%), outperforming conventional Doppler ultrasound at 89.1% (CI: 87.0–91.2%). It reduced false-positive rates by 23% without compromising sensitivity (96.8%), demonstrated robust cross-demographic performance, and introduced the Hemodynamic Complexity Index (HCI), showing an 87% correlation with early vascular disorders. With 38% faster computational efficiency, delivering results in under 15 seconds, AAP offers transformative potential in precision medicine and early cardiovascular care.

AB - Artificial Intelligence (AI) has revolutionized medical diagnostics, notably enabling noninvasive cardiovascular health evaluations. This study introduces Adaptive Acoustic Profiling (AAP), an AI-based approach using a hybrid Convolutional Recurrent Neural Network (CRNN) to detect micro-acoustic blood flow anomalies. Leveraging the public PhysioNet MIMIC-III database, AAP achieved 94.2% diagnostic sensitivity (CI: 92.8–95.6%), outperforming conventional Doppler ultrasound at 89.1% (CI: 87.0–91.2%). It reduced false-positive rates by 23% without compromising sensitivity (96.8%), demonstrated robust cross-demographic performance, and introduced the Hemodynamic Complexity Index (HCI), showing an 87% correlation with early vascular disorders. With 38% faster computational efficiency, delivering results in under 15 seconds, AAP offers transformative potential in precision medicine and early cardiovascular care.

UR - https://www.scopus.com/pages/publications/105025315827

U2 - 10.1201/9781003541011-9

DO - 10.1201/9781003541011-9

M3 - Chapter

AN - SCOPUS:105025315827

SN - 9781032864815

SP - 127

EP - 140

BT - Computer Vision in Healthcare Prediction, Detection and Diagnosis

PB - CRC Press

ER -

A Novel Artificial Intelligence Driven Adaptive Acoustic Profiling Technique for Noninvasive Cardiovascular Diagnostics

Abstract

Bibliographical note

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this