No abstract

No abstract

This paper presents a technique for automated curation of a domain-specific knowledge base or lexicon for resource-constrained domains, such as Emergency Medical Services (EMS) and its application to real-time concept extraction and cognitive assistance in emergency response. The EMS responders often verbalize critical information describing the situations at an incident scene, including patients' physical condition and medical history. Automated extraction of EMS protocol-specific concepts from responders' speech data can facilitate cognitive support through the selection and execution of the proper EMS protocols for patient treatment. Although this task is similar to the traditional NLP task of concept extraction, the underlying application domain poses major challenges, including low training resources availability (e.g., no existing EMS ontology, lexicon, or annotated EMS corpus) and domain mismatch. Hence, we develop EMSContExt, a weakly-supervised concept extraction approach for EMS concepts. It utilizes different knowledge bases and a semantic concept model based on a corpus of over 9400 EMS narratives for lexicon expansion. The expanded EMS lexicon is then used to automatically extract critical EMS protocol-specific concepts from real-time EMS speech narratives. Our experimental results show that EMSContExt achieves 0.85 recall and 0.82 F1-score for EMS concept extraction and significantly outperforms MetaMap, a state-of-the-art medical concept extraction tool. We also demonstrate the application of EMSContExt to EMS protocol selection and execution and real-time recommendation of protocol-specific interventions to the EMS responders. Here, EMSContExt outperforms MetaMap with a 6% increase and six times speedup in weighted recall and execution time, respectively.

This paper presents our preliminary results on development of a Cognitive assistant system for Emergency Medical Services (CognitiveEMS) that aims to improve situational awareness and safety of first responders. CognitiveEMS integrates a suite of smart wearable sensors, devices, and analytics for real-time collection and analysis of in-situ data from incident scene and delivering dynamic data-driven insights to responders on the most effective response actions to take. We present the overall architecture of CognitiveEMS pipeline for processing information collected from the responder, which includes stages for converting speech to text, extracting medical and EMS protocol specific concepts, and modeling and execution of an EMS protocol. The performance of the pipeline is evaluated in both noise-free and noisy incident environments. The experiments are conducted using two types of publicly-available real EMS data: short radio calls and post-incident patient care reports. Three different noise profiles are considered for simulating the noisy environments: cafeteria, people talking, and emergency sirens. Noise was artificially added at 3 intensity levels of low, medium, and high to pre-recorded audio data. The results show that the i) state-of-the-art speech recognition tools such as Google Speech API are quite robust to low and medium noise intensities; ii) in the presence of high noise levels, the overall recall rate in medical concept annotation is reduced; and iii) the effect of noise often propagates to the final decision making stage and results in generating misleading feedback to responders.