RTX BBN Technologies to support ARPA-H AI-powered medical chatbots reliability evaluation effort
RTX BBN Technologies has been awarded a contract to support ARPA-H's Chatbot Accuracy and Reliability Evaluation (CARE) initiative. The company will develop the Monitoring, Evaluation and Diagnosing of Intelligent Chatbots (MEDIC) system to assess medical chatbots' reliability in healthcare settings.
The project aims to address critical limitations in current AI healthcare systems, which often generate inaccurate or misleading responses. The MEDIC framework will include integration of healthcare stakeholder insights, medical text validation, demographic-specific prompt engineering, and advanced information extraction techniques to detect inaccuracies.
The initiative involves collaboration with Johns Hopkins University, Johns Hopkins University School of Medicine, and Howard University Hospital, with work being conducted in Cambridge, Baltimore, and Washington, D.C.
RTX BBN Technologies ha ricevuto un contratto per supportare l'iniziativa di valutazione dell'accuratezza e dell'affidabilità dei chatbot di ARPA-H, denominata CARE. L'azienda svilupperà il sistema di Monitoraggio, Valutazione e Diagnosi dei Chatbot Intelligenti (MEDIC) per valutare l'affidabilità dei chatbot medici negli ambienti sanitari.
Il progetto mira ad affrontare le limitazioni critiche degli attuali sistemi sanitari basati su IA, che spesso generano risposte imprecise o fuorvianti. Il framework MEDIC includerà l'integrazione delle intuizioni degli stakeholder del settore sanitario, la validazione dei testi medici, l'ingegneria dei prompt specifici per la demografia e tecniche avanzate di estrazione delle informazioni per rilevare le imprecisioni.
L'iniziativa prevede una collaborazione con Johns Hopkins University, la Scuola di Medicina della Johns Hopkins University e l'Ospedale Universitario di Howard, con lavori in corso a Cambridge, Baltimore e Washington, D.C.
RTX BBN Technologies ha sido adjudicada un contrato para apoyar la iniciativa de Evaluación de Precisión y Fiabilidad de Chatbots (CARE) de ARPA-H. La empresa desarrollará el sistema de Monitoreo, Evaluación y Diagnóstico de Chatbots Inteligentes (MEDIC) para evaluar la fiabilidad de los chatbots médicos en entornos de salud.
El proyecto tiene como objetivo abordar las limitaciones críticas de los actuales sistemas de IA en salud, que a menudo generan respuestas inexactas o engañosas. El marco MEDIC incluirá la integración de las perspectivas de las partes interesadas en el sector salud, la validación de textos médicos, la ingeniería de mensajes específicos para demografías y técnicas avanzadas de extracción de información para detectar inexactitudes.
La iniciativa implica colaboración con Johns Hopkins University, la Escuela de Medicina de la Johns Hopkins University, y el Hospital Universitario de Howard, con trabajo realizado en Cambridge, Baltimore y Washington, D.C.
RTX BBN Technologies는 ARPA-H의 챗봇 정확성과 신뢰성 평가(CARE) 이니셔티브에 지원하기 위해 계약을 체결했습니다. 이 회사는 의료 환경에서 의료 챗봇의 신뢰성을 평가하기 위한 지능형 챗봇 모니터링, 평가 및 진단 시스템 (MEDIC)을 개발할 예정입니다.
이 프로젝트는 종종 부정확하거나 오해의 소지가 있는 응답을 생성하는 현재의 AI 헬스케어 시스템의 중대한 한계를 해결하는 것을 목표로 하고 있습니다. MEDIC 프레임워크는 헬스케어 이해관계자의 통찰력 통합, 의료 텍스트 검증, 인구통계학적 특정 프롬프트 설계 및 부정확성을 탐지하기 위한 고급 정보 추출 기술을 포함할 것입니다.
이 이니셔티브는 존스 홉킨스 대학교, 존스 홉킨스 의과대학 및 하워드 대학교 병원과 협력하며, 캠브리지, 볼티모어 및 워싱턴 D.C.에서 진행됩니다.
RTX BBN Technologies a remporté un contrat pour soutenir l'initiative d'évaluation de l'exactitude et de la fiabilité des chatbots (CARE) de l'ARPA-H. L'entreprise développera le système de Monitoring, Évaluation et Diagnostic des Chatbots Intelligents (MEDIC) pour évaluer la fiabilité des chatbots médicaux dans les contextes de soins de santé.
Ce projet vise à remédier aux limitations critiques des systèmes de santé basés sur l'IA actuels, qui génèrent souvent des réponses inexactes ou trompeuses. Le cadre MEDIC inclura l'intégration des perspectives des parties prenantes de la santé, la validation des textes médicaux, l'ingénierie de prompts spécifiques à la démographie et des techniques avancées d'extraction d'informations pour détecter les inexactitudes.
L'initiative implique une collaboration avec la Johns Hopkins University, la faculté de médecine de la Johns Hopkins University et l'hôpital de l'université de Howard, avec des travaux effectués à Cambridge, Baltimore et Washington, D.C.
RTX BBN Technologies wurde mit einem Vertrag zur Unterstützung der Initiative zur Bewertung der Genauigkeit und Zuverlässigkeit von Chatbots (CARE) von ARPA-H beauftragt. Das Unternehmen wird das System zur Überwachung, Bewertung und Diagnostik intelligenter Chatbots (MEDIC) entwickeln, um die Zuverlässigkeit medizinischer Chatbots in Gesundheitseinrichtungen zu bewerten.
Das Projekt zielt darauf ab, kritische Einschränkungen der aktuellen KI-Gesundheitssysteme zu adressieren, die oft ungenaue oder irreführende Antworten generieren. Das MEDIC-Framework wird die Integration von Erkenntnissen der Gesundheitsbeteiligten, die Validierung medizinischer Texte, die demografiespezifische Gestaltungsengineering und fortschrittliche Techniken zur Informationsbeschaffung zur Erkennung von Ungenauigkeiten umfassen.
Die Initiative erfolgt in Zusammenarbeit mit der Johns Hopkins University, der medizinischen Fakultät der Johns Hopkins University und dem Howard University Hospital, mit Arbeiten, die in Cambridge, Baltimore und Washington, D.C. durchgeführt werden.
- Secured new government contract from ARPA-H for healthcare AI technology development
- Strategic partnership with prestigious institutions (Johns Hopkins University, Howard University Hospital)
- Expansion into growing healthcare AI market with proprietary MEDIC system
- None.
Insights
The development of MEDIC represents a significant advancement in AI healthcare validation technology. By integrating multiple evaluation methods and real-world medical expertise, this system addresses a critical gap in healthcare technology validation. The involvement of prestigious institutions like Johns Hopkins adds substantial credibility to the project.
The system's ability to validate responses against evidence-based sources while considering diverse demographic perspectives is particularly noteworthy. This comprehensive approach could become an industry standard for medical AI validation, potentially influencing future healthcare technology development and regulation.
However, the project's success will largely depend on its ability to scale and adapt to rapidly evolving AI technologies. While the framework shows promise, its real impact on healthcare technology adoption and patient safety remains to be proven in practical applications.
This initiative addresses a important need in healthcare technology safety. The MEDIC system's focus on validating medical chatbot responses against evidence-based sources could significantly reduce the risk of misinformation in patient care. The integration of multiple stakeholder perspectives - from caregivers to medical professionals - ensures a well-rounded evaluation framework.
The emphasis on detecting bias and fairness issues is particularly important given the diverse patient populations in healthcare. The prenatal care example highlighted demonstrates the system's practical application in high-stakes medical scenarios where accurate information is important for patient safety.
The collaboration between technical experts and medical institutions suggests a robust approach to developing healthcare AI validation standards. This could lead to improved patient trust in AI-powered healthcare tools while maintaining high safety standards.
BBN developing technology to assess the reliability and accuracy of healthcare responses
Despite the potential of medical chatbots, significant limitations threaten their effectiveness. Many AI systems generate factually inaccurate or misleading responses that may cause confusion and pose potential risk to patients. As healthcare evolves, a scalable system is needed to ensure consistent medical chatbot performance in any setting. This need is intensified by ongoing lack of standardization, which continues to undermine confidence.
"Evaluating medical chatbots requires more than simply checking for correct answers; it demands a deep understanding of how these systems address the complex needs of diverse users," said Dr. Damianos Karakos, BBN principal investigator on the effort.
To address this problem, BBN will use its expertise in machine learning, language-based information processing and large language models to develop the Monitoring, Evaluation and Diagnosing of Intelligent Chatbots (MEDIC) system. This comprehensive solution will function as a technological framework for evaluating medical chatbots, featuring core capabilities such as:
- Integration of insights from caregivers, patients and medical professionals to optimize chatbot interactions and effectively address their concerns and expectations.
- Retrieval of relevant medical texts to validate chatbot responses against evidence-based data sources.
- Advanced prompt engineering to create realistic interactions from various demographic perspectives.
- Detection of missing or inaccurate information in chatbot outputs using multiple evaluative methods, which use advanced information extraction and machine learning techniques.
"Our goal is to develop an adaptable framework that rigorously assesses chatbot performance in real-world scenarios, focusing on key aspects like bias, fairness and the risk of generating misleading information," said Karakos. "For example, in prenatal care, it's crucial that expectant mothers receive accurate dietary guidance to support fetal health. MEDIC will assess the dietary advice given by medical chatbots and escalate any ambiguous responses to healthcare professionals for further review. This initiative aims to improve AI-integrated care in a variety of healthcare settings."
The BBN-led team includes Johns Hopkins University (Prof. Mark Dredze), Johns Hopkins University School of Medicine and Howard University Hospital. Work on this effort is being performed in
This research was, in part, funded by the Advanced Research Projects Agency for Health (ARPA-H). The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of the United States Government.
About RTX BBN Technologies
Founded in 1948, RTX BBN Technologies provides advanced technology research and development with a focus on national security priorities. From the ARPANET to the first email, through the first metro network protected by quantum cryptography, BBN consistently transitions advanced research to produce innovative solutions for its customers. BBN takes risks and challenges conventions to create solutions in analytics and machine intelligence, networks and sensors, intelligent software and systems, and physical sciences.
About RTX
With more than 185,000 global employees, RTX pushes the limits of technology and science to redefine how we connect and protect our world. Through industry-leading businesses – Collins Aerospace, Pratt & Whitney, and Raytheon – we are advancing aviation, engineering integrated defense systems for operational success, and developing next-generation technology solutions and manufacturing to help global customers address their most critical challenges. The company, with 2023 sales of
For questions or to schedule an interview, please contact corporatepr@rtx.com
View original content:https://www.prnewswire.com/news-releases/rtx-bbn-technologies-to-support-arpa-h-ai-powered-medical-chatbots-reliability-evaluation-effort-302327161.html
SOURCE RTX
FAQ
What is RTX BBN Technologies developing for ARPA-H's CARE initiative?
Who are RTX's partners in the ARPA-H CARE medical chatbot project?
Where will RTX conduct the ARPA-H CARE project research?
What are the main features of RTX's MEDIC system for medical chatbots?
