Article
Medicine, General & Internal
Chung-Feng Liu, Chao-Ming Hung, Shian-Chin Ko, Kuo-Chen Cheng, Chien-Ming Chao, Mei- Sung, Shu-Chen Hsing, Jhi-Joung Wang, Chia-Jung Chen, Chih-Cheng Lai, Chin-Ming Chen, Chong-Chi Chiu
Summary: Using AI approach, two-stage predictive models were developed to determine the optimal timing for weaning intubated ICU patients from mechanical ventilation. The models were found to effectively predict the timing, reduce patient discomfort, improve medical quality, and lower costs. Furthermore, the AI-assisted system proved beneficial in helping clinicians manage high ventilator demand during the COVID-19 pandemic.
FRONTIERS IN MEDICINE
(2022)
Article
Health Care Sciences & Services
Ming-Yen Lin, Yuan-Ming Chang, Chi-Chun Li, Wen-Cheng Chao
Summary: In this study, an explainable machine learning approach was used to develop a weaning prediction model for critically ill ventilated patients requiring hemodialysis. The XGBoost and GBM models showed better accuracy compared to other models. The discriminative characteristics of six key features used for weaning prediction were demonstrated using SHAP and PDP plots. LIME was utilized to provide explanations of predicted probabilities and associated reasoning for successful weaning on an individual level.
Editorial Material
Anesthesiology
Dominic C. Marshall, Matthieu Komorowski
Summary: Artificial intelligence has the potential to improve clinical decision-making for ventilated patients, but the majority of current studies suffer from methodological bias and are not ready for deployment.
BRITISH JOURNAL OF ANAESTHESIA
(2022)
Review
Medicine, General & Internal
Davide Nicolotti, Silvia Grossi, Francesco Nicolini, Alan Gallingani, Sandra Rossi
Summary: Respiratory weaning after cardiac surgery may be difficult or prolonged in up to 22.7% of patients. Inability to wean from a ventilator within 48 hours after surgery is associated with increased morbidity and mortality. Non-modifiable risk factors include preoperative renal failure, New York Heart Association and Canadian Cardiac Society classes, as well as surgery and cardio-pulmonary bypass time. Monitoring certain parameters during weaning, such as pulmonary artery occlusion pressure and brain-derived natriuretic peptide, can help detect hemodynamic decompensation.
JOURNAL OF CLINICAL MEDICINE
(2023)
Article
Medicine, General & Internal
Ming-Yen Lin, Chi-Chun Li, Pin-Hsiu Lin, Jiun-Long Wang, Ming-Cheng Chan, Chieh-Liang Wu, Wen-Cheng Chao
Summary: An explainable machine learning model was developed to predict successful weaning in patients requiring prolonged mechanical ventilation, using real-world data. The XGBoost and RF models outperformed the logistic regression model in predicting successful weaning, with feature importance stratified by clinical domains.
FRONTIERS IN MEDICINE
(2021)
Article
Engineering, Civil
Zubayer Islam, Mohamed Abdel-Aty, Jorge Ugan
Summary: Signal phasing and timing can be adaptive and actuated, making it difficult to predict future cycle length and phase duration. This study proposes a long short-term memory model to predict cycle length and phase duration up to six cycles ahead. By merging GPS information and signal timing information, key features such as waiting time, approach speed, and acceleration are calculated. The results show a mean absolute error of about 7 seconds for cycle length prediction and 9 seconds for phase prediction.
TRANSPORTATION RESEARCH RECORD
(2023)
Article
Business
Tobias Albrecht, Theresa Maria Rausch, Nicholas Daniel Derra
Summary: This study investigates the capabilities of ML models for intra-daily call center arrivals' forecasting, finding that the random forest algorithm yields the best prediction performances. A methodological walk-through example of model selection process is provided to encourage implementation in practical settings.
JOURNAL OF BUSINESS RESEARCH
(2021)
Article
Health Care Sciences & Services
Wei-Chun Tsai, Chung-Feng Liu, Hung-Jung Lin, Chien-Chin Hsu, Yu-Shan Ma, Chia-Jung Chen, Chien-Cheng Huang, Chia-Chun Chen
Summary: This study developed an AI dashboard for emergency department (ED) to monitor the real-time risk of patients. The ED medical staff found the dashboard to be easy to use, useful, and acceptable.
Article
Computer Science, Artificial Intelligence
Valerio La Gatta, Vincenzo Moscato, Marco Postiglione, Giancarlo Sperli
Summary: In this paper, a novel model-agnostic Explainable AI technique named CASTLE is proposed to provide rule-based explanations based on both the local and global model's workings. The framework has been evaluated on six datasets in terms of temporal efficiency, cluster quality and model significance, showing a 6% increase in interpretability compared to another state-of-the-art technique, Anchors.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Multidisciplinary Sciences
Shadi Ebrahimian, Fatemeh Homayounieh, Marcio A. B. C. Rockenbach, Preetham Putha, Tarun Raj, Ittai Dayan, Bernardo C. Bizzo, Varun Buch, Dufan Wu, Kyungsang Kim, Quanzheng Li, Subba R. Digumarthy, Mannudeep K. Kalra
Summary: The study compared the performance of AI and RALE scores in predicting patient outcomes and the need for mechanical ventilation in COVID-19 pneumonia. Both scores were strongly correlated with patients who had higher scores experiencing higher mortality and need for mechanical ventilation. Improved prediction accuracy was achieved by incorporating additional patient information such as age, gender, WBC count, and oxygen saturation.
SCIENTIFIC REPORTS
(2021)
Article
Clinical Neurology
Jiwon Lee, Se Eun Park, Dajeong Lee, Joo Young Song, Jeehun Lee
Summary: SMA type 1 is a severe condition with early onset and fast progression, often requiring permanent assisted ventilation. Nusinersen is currently the only approved treatment that can increase functional SMN protein levels. A case report demonstrates successful weaning from permanent ventilation with tracheostomy using nusinersen in an infant diagnosed with SMA type 1.
ANNALS OF CLINICAL AND TRANSLATIONAL NEUROLOGY
(2021)
Review
Chemistry, Multidisciplinary
Ady Suwardi, FuKe Wang, Kun Xue, Ming-Yong Han, Peili Teo, Pei Wang, Shijie Wang, Ye Liu, Enyi Ye, Zibiao Li, Xian Jun Loh
Summary: Biomaterials research has historically been hindered by long development periods, but the application of machine learning in materials science has greatly accelerated progress. The combination of machine learning with high-throughput theoretical predictions and experiments has shifted the traditional trial and error paradigm to a data-driven paradigm, which is driving the discovery and application of biomaterials.
ADVANCED MATERIALS
(2022)
Review
Medicine, General & Internal
Gerui Zhang, Lin Luo, Limin Zhang, Zhuo Liu
Summary: Machine Learning (ML) is an algorithm that utilizes big data to learn patterns from observed data for accomplishing specific tasks. In the field of medicine, ML has shown great potential in various aspects such as lung imaging analysis, medical monitoring, and prediction evaluation. It is particularly effective in diagnosing interstitial lung disease.
Review
Computer Science, Artificial Intelligence
Muhammed Rashid, Manasvini Ramakrishnan, Viji Pulikkel Chandran, Siddeshappa Nandish, Sreedharan Nair, Vishal Shanbhag, Girish Thunga
Summary: This systematic review summarizes the literature on the various applications of artificial intelligence (AI) in acute respiratory distress syndrome (ARDS). The results show that AI has been widely used in ARDS for diagnosis, risk stratification, prediction of severity, management, prediction of mortality, and decision making.
ARTIFICIAL INTELLIGENCE IN MEDICINE
(2022)
Review
Anesthesiology
Jack Gallifant, Joe Zhang, Maria Del Pilar Arias Lopez, Tingting Zhu, Luigi Camporota, Leo A. Celi, Federico Formenti
Summary: This study analyzed the application of artificial intelligence (AI) in mechanical ventilation and identified limitations such as limited availability of data sets and code, under-reporting of ethnicity and model calibration, and high risk of bias. Potential solutions were proposed to improve confidence in and translation of this promising approach.
BRITISH JOURNAL OF ANAESTHESIA
(2022)
Article
Nutrition & Dietetics
Chia-Cheng Tseng, Chih-Yen Tu, Chia-Hung Chen, Yao-Tung Wang, Wei-Chih Chen, Pin-Kuei Fu, Chin-Ming Chen, Chih-Cheng Lai, Li-Kuo Kuo, Shih-Chi Ku, Wen-Feng Fang
Summary: This study compared the prognostic accuracy of the modified Nutrition Risk in Critically Ill (mNUTRIC) score with other clinical prediction rules for severe community-acquired pneumonia (SCAP) patients. The results showed that the mNUTRIC score was a better predictor of clinical outcomes compared to other prediction rules.
Article
Infectious Diseases
Chien-An Chen, Chung-Han Ho, Yu-Cih Wu, Yi-Chen Chen, Jhi-Joung Wang, Kuang-Ming Liao
Summary: The incidence of aspergillosis is increasing in cancer patients, and cancer patients with aspergillosis have a significantly higher risk of mortality.
INFECTION AND DRUG RESISTANCE
(2022)
Article
Respiratory System
Kuang-Ming Liao, Yi-Ju Chen, Chuan-Wei Shen, Shao-Kai Ou, Chung-Yu Chen
Summary: This study reviewed the impact of influenza infection on COPD patients and found that influenza infection is associated with increased mortality, pneumonia, respiratory failure, COPD acute exacerbation, and ischemic stroke among COPD patients.
INTERNATIONAL JOURNAL OF CHRONIC OBSTRUCTIVE PULMONARY DISEASE
(2022)
Article
Respiratory System
Kuang-Ming Liao, Jhi-Joung Wang, Chung-Han Ho
Summary: This study aimed to assess the long-term outcomes of COPD patients receiving triple therapy in real-world practice. The study found that over a 5-year observation period, COPD patients receiving triple therapy did not show a survival benefit compared to those not receiving triple therapy.
INTERNATIONAL JOURNAL OF CHRONIC OBSTRUCTIVE PULMONARY DISEASE
(2023)
Article
Oncology
Shou-Chun Yu, Yow-Ling Shiue, Yu-Cih Wu, Jhi-Joung Wang, Kuang-Ming Liao, Chung-Han Ho
Summary: The global incidence of early-onset colorectal cancer (EO-CRC) is increasing, with higher mortality rates associated with certain comorbidities. This study used real-world data to estimate the mortality risk in EO-CRC patients with various comorbidities. The incidence rate of EO-CRC significantly increased from 2007 to 2017. Cerebrovascular disease (CVD) and chronic kidney disease were associated with higher mortality risks, while radiotherapy was linked to higher overall mortality risk.
FRONTIERS IN ONCOLOGY
(2023)
Article
Medicine, General & Internal
Kuang-Ming Liao, Chung-Shu Lee, Yu-Cih Wu, Chin-Chung Shu, Chung-Han Ho
Summary: This study analyzes the mortality risk of lung cancer patients with prior tuberculosis (TB). The results show that lung cancer patients with prior TB have a higher risk of mortality within 3 years after diagnosis.
FRONTIERS IN MEDICINE
(2023)
Article
Medicine, General & Internal
Kuang-Ming Liao, Chung-Feng Liu, Chia-Jung Chen, Jia-Yih Feng, Chin-Chung Shu, Yu-Shan Ma
Summary: This study aimed to build models using an artificial intelligence/machine learning approach to predict acute hepatitis, respiratory failure, and mortality after TB treatment. The results showed that our models using the six AI algorithms all had a high accuracy in predicting these adverse effects.
Article
Public, Environmental & Occupational Health
Cheng-Yi Wang, Kuang-Ming Liao, Ya -Hui Wang, Kuang-Hung Chen, Shulin Chuang, Chia-Jung Liu, Chin-Chung Shu, Hao-Chien Wang
Summary: This study suggests that the use of DPP4i in patients with type 2 diabetes does not increase the risk of mycobacterial pulmonary infections.
JOURNAL OF INFECTION AND PUBLIC HEALTH
(2023)
Review
Medicine, General & Internal
I-Jung Feng, Jia-Wei Lin, Chih-Cheng Lai, Kuo-Chen Cheng, Chin-Ming Chen, Chien-Ming Chao, Ying-Ting Wang, Shyh-Ren Chiang, Kuang-Ming Liao
Summary: In this study, the efficacy of various corticosteroid treatments for preventing postextubation stridor and reintubation in mechanically ventilated adults with planned extubation were assessed. Methylprednisolone and dexamethasone were found to be the most effective agents against postextubation stridor and reintubation.
FRONTIERS IN MEDICINE
(2023)
Article
Oncology
Kuang-Ming Liao, Chin-Chung Shu, Fu-Wen Liang, Yi-Chen Chen, Chia-Hung Yu, Jhi-Joung Wang, Chung-Han Ho
Summary: This study analyzed the risk factors for newly diagnosed pulmonary tuberculosis (PTB) among lung cancer patients in Taiwan. The results showed that age, gender, history of pneumoconiosis, and treatments of surgery and chemotherapy were associated with an increased risk of PTB. Screening for PTB may be important among lung cancer patients with these risk factors.
Article
Oncology
Shou-Chun Yu, Kuang-Ming Liao, Chia-Lin Chou, Yu-Feng Tian, Jhi-Joung Wang, Chung-Han Ho, Yow-Ling Shiue
Summary: This study investigates the mortality differences between different primary tumor locations in colorectal cancer and finds that colon cancers have a worse prognosis and survival in women and older patients.
CLINICAL MEDICINE INSIGHTS-ONCOLOGY
(2022)
Article
Medicine, General & Internal
Hsueh-Yi Lu, Kuang-Ming Liao
Summary: Bronchiectasis is a common comorbidity in chronic obstructive pulmonary disease (COPD). This study used a nationwide database to evaluate the incidence of bronchiectasis in COPD in Taiwan. The results showed that the incidence of bronchiectasis in COPD patients was higher than in non-COPD patients, and the risk increased with age.
Article
Respiratory System
Kuo-Chen Cheng, Chih-Cheng Lai, Cheng-Yi Wang, Ching-Min Wang, Chung-Han Ho, Mei- Sung, Shu-Chen Hsing, Kuang-Ming Liao, Shian-Chin Ko
Summary: This study investigates the impact of a multidisciplinary intervention on the clinical outcomes of patients with COPD. The results show that the intervention can improve pulmonary function and symptom scores in COPD patients, but the decrease in exacerbation frequency was only observed in certain patient groups.
INTERNATIONAL JOURNAL OF CHRONIC OBSTRUCTIVE PULMONARY DISEASE
(2022)