A New Real Time Clinical Decision Support System Using Machine Learning for Critical Care Units

Abstract

Mean arterial pressure (MAP) is an important clinical parameter to evaluate the health of critically ill patients in intensive care units. Thus, the real time clinical decision support systems detecting anomalies and deviations in MAP enable early interventions and prevent serious complications. The state-ofthe-art decision support systems are based on a three-phase method that applies offline training, transfer learning, and retraining at the bedside. Their applicability in critical care units is challenging with delay and inaccuracy. In this paper, we propose a real time clinical decision support system forecasting the MAP status at the bedside using a new machine learning structure. The proposed system works in real time at the bedside without requiring the offline phase for training using large datasets. It thereby enables timely interventions and improved healthcare services. The proposed machine learning structure includes two stages. Stage I applies online learning using hierarchical temporal memory (HTM) to enable real time stream processing and provides unsupervised predictions. To the best of our knowledge, this is the first time it is applied to medical signals. Stage II is a long short-term memory (LSTM) classifier that forecasts the status of the patient's MAP ahead of time based on Stage I stream predictions. We perform a thorough performance evaluation of the proposed system and compare it with the state-of-the-art systems employing logistic regression (LR). The comparison shows the proposed system outperforms LR in terms of the classification accuracy, recall, precision, and area under the receiver operation curve (AUROC). INDEX TERMS Clinical decision support, classification, hierarchical temporal memory (HTM), long shortterm memory (LSTM), machine learning, real time prediction.

Predicting the Mortality of ICU Patients by Topic Model with Machine-Learning Techniques

“…Proposed by Blei et al [ 19 ], LDA is a typical “bag of words” model that treats each text as a vocabulary frequency vector and as a collection of multiple sets of vocabularies. In addition, each group of vocabularies represents a topic, and text topics are extracted without considering the order of and relevance between the vocabularies [ 37 , 38 ]. Normally, an LDA builds its topic generation model through the following steps: (1) a topic is selected from the various topics in a text; (2) a vocabulary is chosen from the list of vocabularies corresponding to the topic selected; and (3) the process is repeated until all of the vocabulary in the text has been selected.…”

Deep Ensemble Learning with GAN-based Semi-Supervised Training Algorithm for Medical Decision Support System in Healthcare Applications

“…CDSS for real-time monitoring of mean arterial pressure (MAP) status was designed [21] at the bedside using a novel machine-learning algorithm. Initially, online learning was applied by hierarchical temporal memory (HTM) to enable real-time stream processing and obtain unsupervised predictions.…”

“…Decision-making assistance unit: The welllearned semi-supervised trainer conducts a categorization process, where an unknown data like a patient is allocated via the tag, which estimates the unknown membership in one of the deliberate labels defining probable prognosis of multiple disorders. , Fuzzy_AHP+ANN [17], PSO-DNN [18], HTM+LSTM [21], DBSCAN+SMOTE-ENN+XGBoost [22] and LightGBM [25] regarding the following metrics:…”

Appositeness of Optimized and Reliable Machine Learning for Healthcare: A Survey

“…Dhief et al [ 17 ] presented an extensive review of IoT frameworks and state-of-art techniques used in healthcare and voice pathology surveillance systems whereas Alhussein et al [ 18 ] investigated the voice abnormality detection system using DL on mobile healthcare frameworks. Researchers and physicians are reviewing numerous approaches to utilize the skill of DL methods for Intensive Care Unit (ICUs) and critically acclaimed concerns [ 19 – 21 ], similarly, Ganainy et al [ 22 ] proposed a real-time consultation system in the clinical context which forecasts the Mean Arterial Pressure (MAP) values’ current status at the ease of bed accessibility using new ML structures. The majority of intelligent applications utilizing customer records have received disappointing results at some point in their performance due to their obsession with metrics [ 23 – 25 ].…”