Article
Computer Science, Information Systems
Guangjian Zeng, Jinhu Zhuang, Haofan Huang, Mu Tian, Yi Gao, Yong Liu, Xiaxia Yu
Summary: The mortality rate in the ICU is a critical metric for assessing hospital clinical quality. Existing methods for risk stratification struggle to capture time sequence information, hindering continuous severity assessment during a patient's hospital stay. To address this, we developed a predictive model that can provide real-time risk predictions throughout a patient's stay. Our proposed model outperformed other machine learning methods in accurately predicting risk of death, enabling physicians to prioritize high-risk patients and anticipate potential complications to reduce ICU mortality.
TSINGHUA SCIENCE AND TECHNOLOGY
(2023)
Article
Computer Science, Artificial Intelligence
Nora El-Rashidy, Tamer Abuhmed, Louai Alarabi, Hazem M. El-Bakry, Samir Abdelrazek, Farman Ali, Shaker El-Sappagh
Summary: Sepsis is a life-threatening disease with difficulties in early identification, but establishing an accurate predictive model is crucial for improving prognosis.
NEURAL COMPUTING & APPLICATIONS
(2022)
Article
Clinical Neurology
Ximing Nie, Yuan Cai, Jingyi Liu, Xiran Liu, Jiahui Zhao, Zhonghua Yang, Miao Wen, Liping Liu
Summary: This study found that the random forest algorithm performed the best in predicting hospital mortality for cerebral hemorrhage patients in intensive care units, showing the highest specificity and accuracy.
FRONTIERS IN NEUROLOGY
(2021)
Article
Physiology
Vanshika Vats, Aditya Nagori, Pradeep Singh, Raman Dutt, Harsh Bandhey, Mahika Wason, Rakesh Lodha, Tavpritesh Sethi
Summary: In this study, a noncontact thermal imaging modality and deep learning were used to predict shock status up to the next 6 h by analyzing the time-series data of temperature difference and heart rate. The approach provides sufficient time to stabilize the patient.
FRONTIERS IN PHYSIOLOGY
(2022)
Article
Health Care Sciences & Services
I. -Min Chiu, Jhu-Yin Cheng, Tien -Yu Chen, Yi-Min Wang, Chi -Yung Cheng, Chia-Te Kung, Fu-Jen Cheng, Fei-Fei Flora Yau, Chun-Hung Richard Lin
Summary: This study aimed to improve the accuracy of hyperkalemia predictions from ambulatory electrocardiogram (ECG) monitors using a personalized transfer learning method. By training a generic model and refining it with personal data, the personalized model significantly improved the accuracy of hyperkalemia detection.
JOURNAL OF MEDICAL INTERNET RESEARCH
(2022)
Article
Computer Science, Information Systems
Subhrajit Roy, Diana Mincu, Eric Loreaux, Anne Mottram, Ivan Protsyuk, Natalie Harris, Yuan Xue, Jessica Schrouff, Hugh Montgomery, Alistair Connell, Nenad Tomasev, Alan Karthikesalingam, Martin Seneviratne
Summary: The SeqSNR architecture demonstrated a modest yet statistically significant performance boost across 4 of the 6 tasks compared to single-task and naive multitasking approaches. When reducing the size of the training dataset for specific tasks, SeqSNR outperformed single-task in all cases, indicating superior label efficiency especially in scenarios where endpoint labels are difficult to ascertain.
JOURNAL OF THE AMERICAN MEDICAL INFORMATICS ASSOCIATION
(2021)
Article
Computer Science, Information Systems
Tingyi Wanyan, Akhil Vaid, Jessica K. De Freitas, Sulaiman Somani, Riccardo Miotto, Girish N. Nadkarni, Ariful Azad, Ying Ding, Benjamin S. Glicksberg
Summary: Traditional machine learning models have had limited success in predicting COVID-19 outcomes using EHR data, but a novel framework based on relational learning and heterogeneous graph model shows improved prediction accuracy. By leveraging diverse patient populations' EHR data in NYC, the model effectively captures patterns in patient representations of outcomes through relational learning strategy, leading to significant improvements in recall rates.
IEEE TRANSACTIONS ON BIG DATA
(2021)
Article
Health Care Sciences & Services
Andre Moser, Matti Reinikainen, Stephan M. Jakob, Tuomas Selander, Ville Pettila, Olli Kiiski, Tero Varpula, Rahul Raj, Jukka Takala
Summary: This study developed and validated a model for predicting in-hospital mortality risk in ICU patients, aiming to facilitate benchmarking, quality assurance, and health economics evaluation. The model showed good internal validity and geographic discrimination transportability, but significant performance heterogeneity was found between ICUs.
JOURNAL OF CLINICAL EPIDEMIOLOGY
(2022)
Article
Medicine, General & Internal
Kumiko Tanaka, Taka-aki Nakada, Nozomi Takahashi, Takahiro Dozono, Yuichiro Yoshimura, Hajime Yokota, Takuro Horikoshi, Toshiya Nakaguchi, Koichiro Shinozaki
Summary: The study demonstrated the advantages of using machine learning technique for X-ray diagnosis in ICU settings based on portable chest radiograph data from a Japanese hospital and the NIH dataset, showing higher diagnostic accuracy and faster processing time compared to ICU physicians.
FRONTIERS IN MEDICINE
(2021)
Article
Health Care Sciences & Services
Kaouter Karboub, Mohamed Tabaa
Summary: This paper addresses the challenge of allocating medical resources in ICUs effectively. It trains regression models using the MIMIC III database and evaluates their performance on unseen data. The best model achieves a high accuracy in predicting patients' readiness for discharge, highlighting the influential factors in making the decision.
Article
Critical Care Medicine
Joao Gabriel Rosa Ramos, Gabriel Machado Naus dos Santos, Marina Chetto Coutinho Bispo, Renata Cristina de Almeida Matos, Gil Mario Lopes Santos de Carvalho Jr, Rogerio da Hora Passos, Juliana Ribeiro Caldas, Andre Luiz Nunes Gobatto, Suzete Nascimento Farias da Guarda, Paulo Benigno Pena Batista
Summary: Unplanned transfers from the intermediate care unit to the intensive care unit were common among urgent admissions, associated with increased hospital mortality, and mostly due to deterioration in the condition that was the reason for admission.
AMERICAN JOURNAL OF CRITICAL CARE
(2021)
Article
Multidisciplinary Sciences
Xiao Chen, Xiaofeng Zhu, Huichang Zhuo, Jiandong Lin, Xian Lin
Summary: This study reveals that the absence of basophils in ICU patients is strongly associated with poor prognosis and mortality. The findings suggest that basophil-mediated immunity may serve as a potential predictor for ICU patients' prognosis, and help identify patients who would benefit from intensified treatment and immunoenhancement therapy.
SCIENTIFIC REPORTS
(2023)
Article
Multidisciplinary Sciences
Min Hyuk Choi, Dokyun Kim, Eui Jun Choi, Yeo Jin Jung, Yong Jun Choi, Jae Hwa Cho, Seok Hoon Jeong
Summary: Improving predictive models for ICU inpatients requires a new strategy that periodically includes the latest clinical data and can be updated to reflect local characteristics. Machine learning algorithms outperformed conventional scoring models in predicting in-hospital mortality. Combining machine learning models with conventional scoring systems can provide more useful information for predicting the prognosis of critically ill patients. Training the predictive model with individual data from each hospital can enhance its robustness.
SCIENTIFIC REPORTS
(2022)
Article
Computer Science, Interdisciplinary Applications
Firuz Juraev, Shaker El-Sappagh, Eldor Abdukhamidov, Farman Ali, Tamer Abuhmed
Summary: This study utilizes advanced machine learning approaches to predict mortality and length of stay for neonates in intensive care units. By building a multilayer dynamic ensemble-based model and optimizing it using real data, the study demonstrates the superiority of dynamic ensemble models in these tasks and supports the model's decisions using explainability techniques.
JOURNAL OF BIOMEDICAL INFORMATICS
(2022)
Article
Environmental Sciences
Jagadish Kumar Mogaraju
Summary: This study used machine learning tools to predict the impact of CO, O3, CH4, and CO2 on deaths from tracheal, bronchus, and lung cancer. Essential features were identified and the best-performing models were selected based on accuracy metrics.
ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH
(2023)
Article
Energy & Fuels
Amir Seifi, Mohammad H. Moradi, Mohamad Abedini, Alireza Jahangiri
JOURNAL OF ENERGY STORAGE
(2020)
Article
Biology
Jamal Esmaelpoor, Mohammad Hassan Moradi, Abdolrahim Kadkhodamohammadi
COMPUTERS IN BIOLOGY AND MEDICINE
(2020)
Article
Energy & Fuels
Mostafa Rezaeimozafar, Mohsen Eskandari, Mohammad Hadi Amini, Mohammad Hasan Moradi, Pierluigi Siano
Article
Engineering, Biomedical
Zahra Ghanbari, Mohammad Hassan Moradi, Alireza Moradi, Jafar Mirzaei
JOURNAL OF MEDICAL AND BIOLOGICAL ENGINEERING
(2020)
Article
Clinical Neurology
Yousef Mohammadi, Mohammad Hassan Moradi
Summary: This study investigated the relationship between brain activity of depression patients and the severity of depression. Significant correlations were found between functional connectivity and complexity measures in different frequency bands and depression severity, especially in the alpha band. The linear regression model demonstrated the potential for accurately predicting depression severity using EEG features from the alpha band.
CLINICAL EEG AND NEUROSCIENCE
(2021)
Article
Biophysics
Jamal Esmaelpoor, Mohammad Hassan Moradi, Abdolrahim Kadkhodamohammadi
Summary: This paper investigates critical aspects of blood pressure monitoring using PPG and ECG, highlighting the importance of ECG waveform for accuracy improvement and the superiority of machine-learned features in performance. The study also reveals that the combination of feature sets does not provide additional information.
PHYSIOLOGICAL MEASUREMENT
(2021)
Article
Engineering, Biomedical
Jamal Esmaelpoor, Zahra Momayez Sanat, Mohammad Hassan Moradi
Summary: This study proposes a clinical set-up using photoplethysmogram signals for continuous blood pressure estimation, achieving better performance with toe PPG compared to earlobe PPG. The algorithm accuracy improves when both signals are applied together. The method provides high estimation consistency and may serve as a possible alternative for inconvenient invasive blood pressure monitoring in clinical environments.
BIOMEDICAL ENGINEERING-BIOMEDIZINISCHE TECHNIK
(2021)
Article
Computer Science, Artificial Intelligence
Monire Sheikh Hosseini, Mahammad Hassan Moradi
Summary: This paper aims to solve the 3D speckle tracking problem through a fuzzy modeling procedure, starting by extracting the suitable local feature descriptor SIFT and aligning relevant features with sets of fuzzy rules. The proposed method achieves an acceptable tracking error below 1 mm and shows potential in discriminating pathological diagnosis from a healthy one through strain analysis.
APPLIED INTELLIGENCE
(2022)
Article
Computer Science, Interdisciplinary Applications
Monire Sheikh Hosseini, Mahammad Hassan Moradi, Mahdi Tabassian, Jan D'hooge
Summary: This study proposes a feature-based non-rigid image registration method using fuzzy rules to estimate cardiac motion on 3D echocardiographic images. Results show that the proposed method is competitive in terms of tracking error and strain estimate accuracy compared to other established registration algorithms.
COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE
(2021)
Article
Engineering, Biomedical
Zahra Tabanfar, Farnaz Ghassemi, Mohammad Hassan Moradi
Summary: This research aims to estimate the brain sources' activities corresponding to SSVEPs signals using the method of Local Fourier Independent Component Analysis. Results showed that the proposed method had acceptable performance for SSVEP source estimation, demonstrating its applicability for stimulus detection and BCI applications.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2022)
Article
Engineering, Biomedical
Zahra Tabanfar, Farnaz Ghassemi, Mohammad Hassan Moradi
Summary: In this study, a subject-independent BCI target detection system based on SSVEP was developed using task-related components and fuzzy memberships. The system achieved good classification results and required only a concise training process compared to other target detection systems.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2023)
Article
Automation & Control Systems
Monire Sheikh Hosseini, Mohammad Hassan Moradi
Summary: This paper introduces a suitable geometric transformation for quantifying cardiac deformation based on a modified fuzzy inference system. The proposed method extracts relevant features of two echocardiographic images to generate proper rules for registration. The results show the competitiveness of the proposed method with the state-of-the-art algorithms and demonstrate its potential for clinical application.
INTERNATIONAL JOURNAL OF FUZZY SYSTEMS
(2023)
Proceedings Paper
Engineering, Electrical & Electronic
Mahdi Taghaddossi, Mohammad Hassan Moradi
Summary: This study investigates effective connectivity changes in the brain when individuals are confronted with different levels of familiarity and desire towards a brand. The study found that watching brands, especially familiar ones, results in stronger effective relations between brain areas.
2021 29TH IRANIAN CONFERENCE ON ELECTRICAL ENGINEERING (ICEE)
(2021)
Article
Engineering, Electrical & Electronic
Rasool Baghbani, Mohammad Behgam Shadmehr, Masoomeh Ashoorirad, Seyyedeh Fatemeh Molaeezadeh, Mohammad Hassan Moradi
Summary: This study introduces a simple and safe method to localize in-depth pulmonary nodules intraoperatively, using a bioimpedance probe with spherical electrodes. By analyzing bioimpedance data, a smart system was designed to differentiate between healthy and tumoral lung tissue with high accuracy. This research shows the feasibility of designing a real-time, safe, and smart system to localize invisible/impalpable pulmonary nodules using the bioimpedance spectrum of lung tissue.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Article
Computer Science, Information Systems
Sahar Khoubani, Mohammad Hassan Moradi
Summary: The paper introduces a fast Frame Rate Up-Conversion method based on Quaternion Wavelet Transform motion estimation, which improves motion estimation accuracy and computational efficiency through three key elements.
MULTIMEDIA TOOLS AND APPLICATIONS
(2021)