When Virtual Assistants Meet Teledermatology: Validation of a Virtual Assistant to Improve the Quality of Life of Psoriatic Patients.

Teledermatology has given dermatologists a tool to track patients' responses to therapy using images. Virtual assistants, the programs that interact with users through text or voice messages, could be used in teledermatology to enhance the interaction of the tool with the patients and healthcare professionals and the overall impact of the medication and quality of life of patients. As such, this work aimed to investigate the effectiveness of using a virtual assistant for teledermatology and its impact on the quality of life. We conducted surveys with the participants and measured the usability of the system with the System Usability Scale (SUS). A total of 34 participants (30 patients diagnosed with moderate-severe psoriasis and 4 healthcare professionals) were included in the study. The measurement of the improvement of quality of life was done by analyzing Psoriasis Quality of Life (PSOLIFE) and Dermatology Life Quality Index (DLQI) questionnaires. The results showed that, on average, the quality of life improved (from 63.8 to 64.8 for PSOLIFE (with a p-value of 0.66 and an effect size of 0.06) and 4.4 to 2.8 for DLQI (with a p-value of 0.04 and an effect size of 0.31)). Patients also used the virtual assistant to do 52 medical consultations. Moreover, the usability is above average, with a SUS score of 70.1. As supported by MMAS-8 results, adherence also improved slightly. Our work demonstrates the improvement of the quality of life with the use of a virtual assistant in teledermatology, which could be attributed to the sense of security or peace of mind the patients get as they can contact their dermatologists directly within the virtual assistant-integrated system.

A Study on the Impacts of Slot Types and Training Data on Joint Natural Language Understanding in a Spanish Medication Management Assistant Scenario.

This study evaluates the impacts of slot tagging and training data length on joint natural language understanding (NLU) models for medication management scenarios using chatbots in Spanish. In this study, we define the intents (purposes of the sentences) for medication management scenarios and two types of slot tags. For training the model, we generated four datasets, combining long/short sentences with long/short slots, while for testing, we collect the data from real interactions of users with a chatbot. For the comparative analysis, we chose six joint NLU models (SlotRefine, stack-propagation framework, SF-ID network, capsule-NLU, slot-gated modeling, and a joint SLU-LM model) from the literature. The results show that the best performance (with a sentence-level semantic accuracy of 68.6%, an F1-score of 76.4% for slot filling, and an accuracy of 79.3% for intent detection) is achieved using short sentences and short slots. Our results suggest that joint NLU models trained with short slots yield better results than those trained with long slots for the slot filling task. The results also indicate that short slots could be a better choice for the dialog system because of their simplicity. Importantly, the work demonstrates that the performance of the joint NLU models can be improved by selecting the correct slot configuration according to the usage scenario.

Validation of a Virtual Assistant for Improving Medication Adherence in Patients with Comorbid Type 2 Diabetes Mellitus and Depressive Disorder.

Virtual assistants are programs that interact with users through text or voice messages simulating a human-based conversation. The development of healthcare virtual assistants that use messaging platforms is rapidly increasing. Still, there is a lack of validation of these assistants. In particular, this work aimed to validate the effectiveness of a healthcare virtual assistant, integrated within messaging platforms, with the aim of improving medication adherence in patients with comorbid type 2 diabetes mellitus and depressive disorder. For this purpose, a nine-month pilot study was designed and subsequently conducted. The virtual assistant reminds patients about their medication and provides healthcare professionals with the ability to monitor their patients. We analyzed the medication possession ratio (MPR), measured the level of glycosylated hemoglobin (HbA1c), and obtained the patient health questionnaire (PHQ-9) score in the patients before and after the study. We also conducted interviews with all participants. A total of thirteen patients and five nurses used and evaluated the proposed virtual assistant using the messaging platform Signal. Results showed that on average, the medication adherence improved. In the final interview, 69% of the patients agreed with the idea of continuing to use the virtual assistant after the study.

Microservice chatbot architecture for chronic patient support.

Chatbots are able to provide support to patients suffering from very different conditions. Patients with chronic diseases or comorbidities could benefit the most from chatbots which can keep track of their condition, provide specific information, encourage adherence to medication, etc. To perform these functions, chatbots need a suitable underlying software architecture. In this paper, we introduce a chatbot architecture for chronic patient support grounded on three pillars: scalability by means of microservices, standard data sharing models through HL7 FHIR and standard conversation modeling using AIML. We also propose an innovative automation mechanism to convert FHIR resources into AIML files, thus facilitating the interaction and data gathering of medical and personal information that ends up in patient health records. To align the way people interact with each other using messaging platforms with the chatbot architecture, we propose these very same channels for the chatbot-patient interaction, paying special attention to security and privacy issues. Finally, we present a monitored-data study performed in different chronic diseases, and we present a prototype implementation tailored for one specific chronic disease, psoriasis, showing how this new architecture allows the change, the addition or the improvement of different parts of the chatbot in a dynamic and flexible way, providing a substantial improvement in the development of chatbots used as virtual assistants for chronic patients.

Validation of the Apple Watch for Heart Rate Variability Measurements during Relax and Mental Stress in Healthy Subjects.

Heart rate variability (HRV) analysis is a noninvasive tool widely used to assess autonomic nervous system state. The market for wearable devices that measure the heart rate has grown exponentially, as well as their potential use for healthcare and wellbeing applications. Still, there is a lack of validation of these devices. In particular, this work aims to validate the Apple Watch in terms of HRV derived from the RR interval series provided by the device, both in temporal (HRM (mean heart rate), SDNN, RMSSD and pNN50) and frequency (low and high frequency powers, LF and HF) domain. For this purpose, a database of 20 healthy volunteers subjected to relax and a mild cognitive stress was used. First, RR interval series provided by Apple Watch were validated using as reference the RR interval series provided by a Polar H7 using Bland-Altman plots and reliability and agreement coefficients. Then, HRV parameters derived from both RR interval series were compared and their ability to identify autonomic nervous system (ANS) response to mild cognitive stress was studied. Apple Watch measurements presented very good reliability and agreement (>0.9). RR interval series provided by Apple Watch contain gaps due to missing RR interval values (on average, 5 gaps per recording, lasting 6.5 s per gap). Temporal HRV indices were not significantly affected by the gaps. However, they produced a significant decrease in the LF and HF power. Despite these differences, HRV indices derived from the Apple Watch RR interval series were able to reflect changes induced by a mild mental stress, showing a significant decrease of HF power as well as RMSSD in stress with respect to relax, suggesting the potential use of HRV measurements derived from Apple Watch for stress monitoring.