RESUMEN
We propose a cognitive system for patient-centric care that leverages and combines natural language processing, semantics, and learning from users over time to support care professionals working with large volumes of patient notes. The proposed methods highlight the entities embedded in the unstructured data to provide a holistic semantic view of an individual. A user-based evaluation is presented, showing consensus between the users and the system.
Asunto(s)
Inteligencia Artificial , Procesamiento de Lenguaje Natural , Semántica , HumanosRESUMEN
This analysis examined the efficacy of an automated postdischarge phone assessment for reducing hospital readmissions. All patients discharged between April 1, 2013, and January 31, 2014, from a single Level 1 trauma hospital of a large regional health system center utilizing an automated postdischarge phone assessment service were contacted via automated call between 24 and 72 hours post discharge. Patients answered 5 questions assessing perceived well-being, understanding of discharge instructions and medication regimen, satisfaction, and scheduled follow-up appointments. Responses could automatically prompt health personnel to speak directly with the patient. Data analysis examined rates of hospital readmission-any admission occurring within 30 days of a previous admission-for 3 broad categories of respondents: Answering Machine, Live Answer, and Unsuccessful. There were 6867 discharges included in the analysis. Of the Live Answer patients, 3035 answered all assessment questions; 153 (5.0%) of these had a subsequent readmission. Of the 738 Unsuccessful patients, 62 (8.4%) had a subsequent readmission. Unsuccessful patients were almost 2 times more likely to have a readmission than those who answered all 5 assessment questions. Of the latter group, readmission rates were highest for those who perceived a worsening of their condition (7.4%), and lowest for those reporting no follow-up appointment scheduled (3.8%). (Population Health Management 2016;19:120-124).
Asunto(s)
Estado de Salud , Readmisión del Paciente , Telecomunicaciones , Adulto , Anciano , Femenino , Humanos , Masculino , Persona de Mediana Edad , Alta del Paciente , Adulto JovenRESUMEN
PURPOSE: To determine whether a wire-guided single drill protocol could be utilized without causing an increase in bone temperature beyond those seen with the traditional techniques of sequential drilling with and without a drill guide. MATERIALS AND METHODS: A bovine femoral bone model was used with thermocouples and infrared temperature measurements to record thermal increase of the bone and drills during implant site preparation. Two thermocouples, one on each side of the osteotomy, were placed 1 mm from the outer diameter of the final drill. Drilling was performed at a constant speed (2,100 rpm) and pressure (2 kg) under continuous room temperature irrigation. Infrared temperature measurements of each drill were taken immediately before and after drilling. The six study groups included standard sequential drilling protocols for 3.5-mm and 4.2-mm final drills with and without the use of a surgical guide, and cannulated single drill technique for 3.5-mm and 4.2-mm drills. Statistical analysis was performed using a Tukey post hoc one-way ANOVA test. P<.05 was determined to be significant. RESULTS: No significant difference in thermal increase was found between single drill cannulated implant site preparation and sequential drilling with or without the use of a drill guide for the 3.5-mm or 4.2-mm drilling sequences, respectively. The thermal increase was found to be significantly less for the 4.2-mm single drill compared with the 3.5-mm sequential drill with surgical guide (P=.046). Infrared temperature measurement revealed no significant difference in drill temperatures throughout the study. CONCLUSIONS: Cannulated single drill technique does not cause an increase in bone temperature greater than that seen with standard sequential drilling with or without a surgical guide.