What You Should Know:
– TytoCare, a pioneering virtual care company dedicated to delivering accessible, high-quality primary care from the comfort of home announced the company has become the first in the world to receive FDA clearance for its AI-based detection of all three major abnormal lung sounds.
– The milestone was reached with the FDA clearance of Tyto Insights™ for Rhonchi Detection, completing TytoCare’s comprehensive AI-powered lung sound suite. With this latest clearance, TytoCare’s innovative platform now offers AI-driven analysis of the three most clinically significant abnormal lung sounds—wheezes, crackles, and rhonchi—seamlessly integrated into both its Home Smart Clinic and Pro Smart Clinic solutions.
Tyto Insights™ for Lung Sounds
Tyto Insights™ for Lung Sounds now stands as the first and only solution globally to receive FDA clearance for AI-based detection of all three major abnormal lung sounds: wheezes (cleared in 2023), crackles (cleared in 2024), and now rhonchi. This achievement sets a new benchmark for virtual respiratory care.
Tyto Insights™ for Lung Sounds is built upon TytoCare’s extensive proprietary database of 1.8 million meticulously recorded lung sounds. The AI leverages sophisticated machine-learning algorithms that have been rigorously trained and validated by leading pulmonologists. Each component of the suite analyzes lung acoustic signals in real-time and intelligently flags potential abnormalities for thorough clinical review. Notably, the Rhonchi detection capability has received clearance as an Rx device and for direct use by end-users.
Clinical Validation Studies
Rigorous clinical validation studies demonstrated that the AI-powered rhonchi detection performs equivalently to, or even better than, non-pulmonologist clinicians. This robust finding confirms the algorithm’s power and reliability in accurately identifying rhonchi. The overall performance and accuracy of the entire Tyto Insights™ software suite for detecting wheeze, crackles, and rhonchi were evaluated using retrospective validation datasets derived from real-world usage of the FDA-cleared compatible Tyto Stethoscope.
The algorithm exhibited strong discriminative ability across all three distinct sound types, as measured by the Area Under the Receiver Operating Characteristic Curve (AUC). Impressively, the algorithm achieved an AUC of 95.85%, significantly outperforming general practitioners (AUC of 79.45%) and even exceeding the performance of experts (AUC of 83.03%). While a range of performance indicators were assessed, the reported statistics specifically highlight the AUC values, which serve as a comprehensive overall performance measure evaluating both positive and negative detections and reflecting the level of confidence clinicians can have in the AI’s results.
“Completing our AI lung sound suite is a turning point for virtual care,” said Dedi Gilad, CEO and Co-Founder of TytoCare. “Wheezes, crackles, and rhonchi are critical markers in diagnosing respiratory conditions, and now our AI assists clinicians in detecting all three using TytoCare during a virtual exam. This enhances decision-making and ensures that patients receive the right care at the right time.”
