AI-based approach for transcribing and classifying unstructured emergency call data: A methodological proposal.

Emergency care-sensitive conditions (ECSCs) require rapid identification and treatment and are responsible for over half of all deaths worldwide. Prehospital emergency care (PEC) can provide rapid treatment and access to definitive care for many ECSCs and can reduce mortality in several different settings. The objective of this study is to propose a method for using artificial intelligence (AI) and machine learning (ML) to transcribe audio, extract, and classify unstructured emergency call data in the Serviço de Atendimento Móvel de Urgência (SAMU) system in southern Brazil. The study used all "1-9-2" calls received in 2019 by the SAMU Novo Norte Emergency Regulation Center (ERC) call center in Maringá, in the Brazilian state of Paraná. The calls were processed through a pipeline using machine learning algorithms, including Automatic Speech Recognition (ASR) models for transcription of audio calls in Portuguese, and a Natural Language Understanding (NLU) classification model. The pipeline was trained and validated using a dataset of labeled calls, which were manually classified by medical students using LabelStudio. The results showed that the AI model was able to accurately transcribe the audio with a Word Error Rate of 42.12% using Wav2Vec 2.0 for ASR transcription of audio calls in Portuguese. Additionally, the NLU classification model had an accuracy of 73.9% in classifying the calls into different categories in a validation subset. The study found that using AI to categorize emergency calls in low- and middle-income countries is largely unexplored, and the applicability of conventional open-source ML models trained on English language datasets is unclear for non-English speaking countries. The study concludes that AI can be used to transcribe audio and extract and classify unstructured emergency call data in an emergency system in southern Brazil as an initial step towards developing a decision-making support tool.

Identification of new putative inhibitors of Mycobacterium tuberculosis 3-dehydroshikimate dehydratase from a combination of ligand- and structure-based and deep learning in silico approaches.

The development of new drugs against Mycobacterium tuberculosis is an essential strategy for fighting drug resistance. Although 3-dehydroquinate dehydratase (MtDHQ) is known to be a highly relevant target for M. tuberculosis, current research shows new putative inhibitors of MtDHQ selected by a large-scale ensemble-docking strategy combining ligand- and target-based chemoinformatic methods to deep learning. Initial chemical library was reduced from 216 million to approximately 460 thousand after pharmacophore, toxicity and molecular weight filters. Final library was subjected to an ensemble-docking protocol in GOLD which selected the top 300 molecules (GHITS). GHITS displayed different structures and characteristics when compared to known inhibitors (KINH). GHITS were further screened by post-docking analysis in AMMOS2 and deep learning virtual screening in DeepPurpose. DeepPurpose predicted that a number of GHITS had comparable or better affinity for the target than KINH. The best molecule was selected by consensus ranking using GOLD, AMMOS2 and DeepPurpose scores. Molecular dynamics revealed that the top hit displayed consistent and stable binding to MtDHQ, making strong interactions with active-site loop residues. Results forward new putative inhibitors of MtDHQ and reinforce the potential application of artificial intelligence methods for drug design. This work represents the first step in the validation of these molecules as inhibitors of MtDHQ.

Binding of piperine to mycobacterial RNA polymerase improves the efficacy of rifampicin activity against Mycobacterium leprae and nontuberculous mycobacteria.

Piperine (PPN) is a known inhibitor of efflux pumps in Mycobacterium tuberculosis and in vitro synergism with rifampicin (RIF) has been proven. The current study evaluates the activity of PPN and synergism with RIF in rapidly and slowly growing nontuberculous mycobacteria (NTM). Also, to propose a possible mechanism of interaction of PPN with M. leprae (Mlp) RNA polymerase (RNAp). Minimal inhibitory concentration and drug combination assay was determined by resazurin microtiter assay and resazurin drug combination assay, respectively. In silico evaluation of PPN binding was performed by molecular docking and molecular dynamics (MD). PPN showed higher antimicrobial activity against rapidly growing NTM (32-128 mg/L) rather than for slowly growing NTM (≥ 256 mg/L). Further, 77.8% of NTM tested exhibited FICI ≤ 0.5 when exposed to PPN and RIF combination, regardless of growth speed. Docking and MD simulations showed a possible PPN binding site at the interface between ß and ß' subunits of RNAp, in close proximity to the trigger-helix and bridge-helix elements. MD results indicated that PPN binding hindered the mobility of these elements, which are essential for RNA transcription. We hypothesize that PPN binding might affect mycobacterial RNAp activity, and, possibly, RIF activity and that this mechanism is partially responsible for synergic behaviors with RIF reported in vitro. Communicated by Ramaswamy H. Sarma.