Review
Cardiac & Cardiovascular Systems
Simrat Gill, Karina Bunting, Claudio Sartini, Victor Roth Cardoso, Narges Ghoreishi, Hae-Won Uh, John A. Williams, Kiliana Suzart-Woischnik, Amitava Banerjee, Folkert W. Asselbergs, Mjc Eijkemans, Georgios Gkoutos, Dipak Kotecha
Summary: This study aimed to assess the diagnostic accuracy of smartphone camera photoplethysmography (PPG) compared with conventional electrocardiogram (ECG) for atrial fibrillation (AF) detection. The results showed that PPG had high sensitivity and specificity for AF detection. However, the quality of the existing studies was low and biased, and further independent research is needed to evaluate the true value of PPG technology in AF detection.
Article
Biology
Miguel Rodrigo, Mahmood Alhusseini, Albert J. Rogers, Chayakrit Krittanawong, Sumiran Thakur, Ruibin Feng, Prasanth Ganesan, Sanjiv M. Narayan
Summary: This study used deep learning to analyze electrogram features and accurately classify atrial fibrillation (AF) and atrial tachycardia (AT). The results showed that deep learning could explain the diagnosis by controlling variations in shape, rate, and timing, and identify specific electrogram shapes as the fingerprint of AF. This research is important for further investigating differences in features among subpopulations of AF patients.
COMPUTERS IN BIOLOGY AND MEDICINE
(2022)
Article
Engineering, Biomedical
Neha, H. K. Sardana, R. Kanawade, N. Dogra
Summary: Photoplethysmography (PPG) is a non-invasive optical technique used for detecting cardiovascular diseases. Researchers propose a new set of morphological features for automated detection of multiple arrhythmias using rule-based and statistical learning-based approaches. The proposed methods are implemented and validated on retrospective and prospective datasets, and show comparable accuracy rates of 98.43%/94.16% (retrospective) and 94.16%/93% (prospective) for rule-based and statistical learning approaches, respectively.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2023)
Article
Medicine, General & Internal
Justin Chu, Wen-Tse Yang, Yao-Ting Chang, Fu-Liang Yang
Summary: This study developed an automated detection method for atrial fibrillation (AFib) based on photoplethysmogram (PPG) signals and visualized the results for reassessment. The method showed high sensitivity, specificity, accuracy, and precision across 460 samples. The findings demonstrate the potential of this user-friendly technology for in-house AFib diagnostics.
Article
Computer Science, Artificial Intelligence
Fahimeh Mohagheghian, Dong Han, Om Ghetia, Darren Chen, Andrew Peitzsch, Nishat Nishita, Eric Y. Ding, Edith Mensah Otabil, Kamran Noorishirazi, Alexander Hamel, Emily L. Dickson, Danielle Dimezza, Khanh-Van Tran, David D. Mcmanus, Ki H. Chon
Summary: This study utilized data collected from smartwatches to improve the detection performance of atrial fibrillation using a denoising autoencoder. The results showed significant improvement in detecting occult AF and increased the amount of analyzable data.
EXPERT SYSTEMS WITH APPLICATIONS
(2024)
Article
Cardiac & Cardiovascular Systems
Savannah F. Bifulco, Fima Macheret, Griffin D. Scott, Nazem Akoum, Patrick M. Boyle
Summary: This study used computational simulations and machine learning to investigate the interaction between ablation-induced scar and existing fibrosis in patients with persistent atrial fibrillation. It found that this interaction plays a key role in recurrent arrhythmia.
JOURNAL OF THE AMERICAN HEART ASSOCIATION
(2023)
Article
Physiology
Eemu-Samuli Valiaho, Pekka Kuoppa, Jukka A. Lipponen, Juha E. K. Hartikainen, Helena Jantti, Tuomas T. Rissanen, Indrek Kolk, Hanna Pohjantahti-Maaroos, Maaret Castren, Jari Halonen, Mika P. Tarvainen, Onni E. Santala, Tero J. Martikainen
Summary: This study evaluated ten robust photoplethysmography features for detection of atrial fibrillation and found that autocorrelation was the most significant feature, reliably detecting atrial fibrillation without the need of pulse detection. Autocorrelation combined with pulse wave morphology-based features could provide a computationally effective and reliable wearable monitoring method in screening of atrial fibrillation.
FRONTIERS IN PHYSIOLOGY
(2021)
Article
Health Care Sciences & Services
Yung-Chuan Huang, Yu-Chen Cheng, Mao-Jhen Jhou, Mingchih Chen, Chi-Jie Lu
Summary: Our study aims to develop an effective machine learning scheme to predict vascular events and bleeding in patients with nonvalvular atrial fibrillation taking dabigatran and identify important risk factors. The results show that random forest and extreme gradient boosting have better performance in predicting vascular events and bleeding.
JOURNAL OF PERSONALIZED MEDICINE
(2022)
Article
Medicine, General & Internal
Amalia Ioanna Moula, Iris Parrini, Cecilia Tetta, Fabiana Luca, Gianmarco Parise, Carmelo Massimiliano Rao, Emanuela Mauro, Orlando Parise, Francesco Matteucci, Michele Massimo Gulizia, Mark La Meir, Sandro Gelsomino
Summary: Atrial fibrillation (AF) is the most common arrhythmia, with its incidence increasing with age and comorbidities. Obstructive sleep apnea (OSA) is a chronic sleep disorder more commonly found in older men. Previous studies have shown a link between OSA and AF, although the prevalence of OSA in AF patients remains unknown due to underdiagnosis. This meta-analysis investigated the association between OSA and AF, using data from 54,271 patients. A strong link was found between these two disorders, with the incidence of AF being 88% higher in patients with OSA. Age and hypertension were found to independently strengthen this association, indicating that treating OSA may help reduce AF recurrence. Further research is needed to confirm these findings.
JOURNAL OF CLINICAL MEDICINE
(2022)
Review
Engineering, Biomedical
Ali Rizwan, Ahmed Zoha, Ismail Ben Mabrouk, Hani M. Sabbour, Ameena Saad Al-Sumaiti, Akram Alomainy, Muhammad Ali Imran, Qammer H. Abbasi
Summary: This paper reviews state-of-the-art ECG data-based machine learning models and signal processing techniques for auto diagnosis of AF, discussing key biomarkers, data collection methods, and modern sensing technologies. Key challenges associated with the development of auto diagnosis solutions for AF are highlighted.
IEEE REVIEWS IN BIOMEDICAL ENGINEERING
(2021)
Review
Biochemistry & Molecular Biology
Baptiste Maille, Nathalie Lalevee, Marion Marlinge, Juliette Vahdat, Giovanna Mottola, Clara Degioanni, Lucille De Maria, Victor Klein, Franck Thuny, Frederic Franceschi, Jean-Claude Deharo, Regis Guieu, Julien Fromonot
Summary: This narrative review discusses the potential role of the adenosinergic system in the pathophysiology of atrial fibrillation (AF). It provides insights into adenosinergic system signaling and its interactions with AF. Activation of adenosine receptors can affect the occurrence and maintenance of AF, and the adenosinergic system is also associated with the modulation of the autonomic nervous system and AF risk factors.
Article
Computer Science, Information Systems
Li Zhu, Viswam Nathan, Jilong Kuang, Jacob Kim, Robert Avram, Jeffrey Olgin, Jun Gao
Summary: Atrial Fibrillation (AF) is a significant cardiac rhythm disorder that can be detected and monitored using wearable sensors. This study developed a highly sensitive and specific AF detection algorithm, and comprehensively validated its performance in a real-world population. The research also demonstrated the robustness and accuracy of the algorithm in daily living scenarios.
IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS
(2022)
Article
Chemistry, Physical
Yuan Xi, Sijing Cheng, Shengyu Chao, Yiran Hu, Minsi Cai, Yang Zou, Zhuo Liu, Wei Hua, Puchuan Tan, Yubo Fan, Zhou Li
Summary: We reported a wearable atrial fibrillation prediction wristband (AFPW) that provides long-term monitoring and diagnosis. AFPW has enhanced signal, strong signal-to-noise ratio, and wireless transmission function. After analyzing and testing a sample library of 385 normal people/patients using linear discriminant analysis, the diagnostic success rate of atrial fibrillation was 91%. These excellent performances demonstrate the great potential of AFPW in wearable device diagnosis and intelligent medical treatment.
Article
Medical Informatics
Zhi Li, Kevin M. Wheelock, Sangeeta Lathkar-Pradhan, Hakan Oral, Daniel J. Clauw, Pujitha Gunaratne, Jonathan Gryak, Kayvan Najarian, Brahmajee K. Nallamothu, Hamid Ghanbari
Summary: A machine learning algorithm was developed to predict episodes of atrial fibrillation (AF) associated with low activity levels, with the potential for clinical applications.
BMC MEDICAL INFORMATICS AND DECISION MAKING
(2021)
Article
Cardiac & Cardiovascular Systems
Eiichi Watanabe, Shunsuke Noyama, Ken Kiyono, Hiroshi Inoue, Hirotsugu Atarashi, Ken Okumura, Takeshi Yamashita, Gregory Y. H. Lip, Eitaro Kodani, Hideki Origasa
Summary: The study compared the performance of random forest, logistic regression, and conventional risk schemes in predicting outcomes of atrial fibrillation patients, finding that the random forest model performed as well as or better than the other models in certain aspects.
CLINICAL CARDIOLOGY
(2021)
Article
Health Care Sciences & Services
Hoon Ko, Jimi Huh, Kyung Won Kim, Heewon Chung, Yousun Ko, Jai Keun Kim, Jei Hee Lee, Jinseok Lee
Summary: In this study, an AI algorithm based on deep learning was developed to automatically detect and quantify ascites. The algorithm segmented abdominopelvic CT images and detected ascites by classifying images. The algorithm achieved high accuracy in both detection and segmentation, providing excellent performance for ascites detection and quantification.
JOURNAL OF MEDICAL INTERNET RESEARCH
(2022)
Article
Multidisciplinary Sciences
Hooseok Lee, Hoon Ko, Heewon Chung, Yunyoung Nam, Sangjin Hong, Jinseok Lee
Summary: In this study, PPGI sensors were mounted on a robot for active and autonomous HR estimation. A proposed algorithm simplified the extraction of facial skin images and selected pixels based on the most frequent saturation value to achieve accurate HR assessment. The algorithm was validated on two datasets, demonstrating high accuracy and processing efficiency.
SCIENTIFIC REPORTS
(2022)
Article
Mathematical & Computational Biology
Marriam Nawaz, Tahira Nazir, Muhammad Attique Khan, Majed Alhaisoni, Jung-Yeon Kim, Yunyoung Nam
Summary: Melanoma is a deadly form of skin cancer that is challenging to accurately identify due to differences in skin lesions and difficulties caused by noise and blurring in images. This research proposes a deep learning model that utilizes object detection and clustering methods to detect and segment melanoma moles. Experimental results demonstrate the effectiveness of this approach in accurately segmenting the lesions.
COMPUTATIONAL AND MATHEMATICAL METHODS IN MEDICINE
(2022)
Article
Environmental Sciences
Yu Jin Jeun, Yunyoung Nam, Seong A. Lee, Jin-Hyuck Park
Summary: This study investigates the effect of customized cognitive training on neural efficiency by analyzing prefrontal cortex activity. The results show that personalized cognitive training can effectively improve executive function and neural efficiency.
INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH
(2022)
Letter
Allergy
Min Ji Koo, Rosie Kwon, Seung Won Lee, Yong Sung Choi, Youn Ho Shin, Sang Youl Rhee, Chanyang Min, Seong Ho Cho, Stephen Turner, So Young Kim, Jinseok Lee, Seung-Geun Yeo, Katrina Abuabara, Young Joo Lee, Jae Il Shin, Jung-Hyun Kim, Jung U. Shin, Dong Keon Yon, Nikolaos G. Papadopoulos
Article
Virology
Heewon Chung, Hoon Ko, Hooseok Lee, Dong Keon Yon, Won Hee Lee, Tae-Seong Kim, Kyung Won Kim, Jinseok Lee
Summary: This study aimed to develop a deep learning model using heart rate data from a smartwatch to diagnose COVID-19 before the onset of symptoms. The model showed high accuracy in diagnosing COVID-19 patients, especially in the unvaccinated group. However, the diagnostic performance decreased in vaccinated patients.
JOURNAL OF MEDICAL VIROLOGY
(2023)
Article
Health Care Sciences & Services
Woocheol Jang, Yong Sung Choi, Ji Yoo Kim, Dong Keon Yon, Young Joo Lee, Sung-Hoon Chung, Chae Young Kim, Seung Geun Yeo, Jinseok Lee
Summary: The study aimed to develop an artificial intelligence model to predict respiratory distress syndrome (RDS) in premature infants to avoid unnecessary treatment. A total of 13,087 very low birth weight infants were assessed using various factors, and a 5-layer deep neural network was proposed for enhanced prediction performance. The model achieved high sensitivity, specificity, accuracy, and area under the curve, and a web application was deployed for easy access to the prediction of RDS in premature infants.
JOURNAL OF MEDICAL INTERNET RESEARCH
(2023)
Review
Virology
Hyunju Yon, Hyoin Shin, Jae Il Shin, Jung U. Shin, Youn Ho Shin, Jinseok Lee, Sang Youl Rhee, Ai Koyanagi, Louis Jacob, Lee Smith, Seung Won Lee, Masoud Rahmati, Suhana Ahmad, Wonyoung Cho, Dong Keon Yon
Summary: A systematic review and meta-analysis of monkeypox infection revealed common clinical features including rash, chills, fever, and lymphadenopathy. This study provides important data on the pathophysiology and epidemiology of monkeypox infections.
REVIEWS IN MEDICAL VIROLOGY
(2023)
Article
Geriatrics & Gerontology
Heewon Chung, Yousun Ko, In-Seob Lee, Hoon Hur, Jimi Huh, Sang-Uk Han, Kyung Won Kim, Jinseok Lee
Summary: This study developed a personalized prognostic artificial intelligence (AI) model to predict 5 year survival at 1 year after gastric cancer surgery based on a large dataset. The AI model showed good performance in both internal and external validation, and the nutritional and fat/muscle indices contributed to the prediction performance.
JOURNAL OF CACHEXIA SARCOPENIA AND MUSCLE
(2023)
Article
Medicine, General & Internal
Gang Won Choi, Dong Keon Yon, Yong Sung Choi, Jinseok Lee, Ki Ho Park, Young Ju Lee, Dong Choon Park, Sang Hoon Kim, Jae Young Byun, Seung Geun Yeo
Summary: Contrary to expectations, the study found no differences in the clinical features or prognosis of Bell's palsy cases during the COVID-19 pandemic compared to those occurring before COVID-19.
JOURNAL OF CLINICAL MEDICINE
(2023)
Article
Chemistry, Analytical
Simone Valenti, Gabriele Volpes, Antonino Parisi, Daniele Peri, Jinseok Lee, Luca Faes, Alessandro Busacca, Riccardo Pernice
Summary: In this study, a novel wearable multisensor ring-shaped probe has been developed to simultaneously acquire photoplethysmographic (PPG) and galvanic skin response (GSR) signals in real-time. The device integrates both the PPG and GSR sensors onto a single probe placed on the finger, enabling the extraction of various physiological indices and dynamic changes over time.
Article
Medicine, General & Internal
Abhilash Pati, Manoranjan Parhi, Binod Kumar Pattanayak, Debabrata Singh, Vijendra Singh, Seifedine Kadry, Yunyoung Nam, Byeong-Gwon Kang
Summary: Women across all countries are at the highest risk of breast cancer. Early diagnosis and staging can improve treatment outcomes. Technology enables automatic analysis of medical images, while IoT is crucial for early and remote diagnosis. This study trained a deep transfer learning model using mammography images for autonomous breast cancer diagnosis.
Review
Oncology
Kyung Won Kim, Jimi Huh, Bushra Urooj, Jeongjin Lee, Jinseok Lee, In-Seob Lee, Hyesun Park, Seongwon Na, Yousun Ko
Summary: Gastric cancer is a major global health problem that requires advancements in imaging techniques to ensure accurate diagnosis and effective treatment planning. Artificial intelligence has emerged as a powerful tool for gastric cancer imaging, particularly in diagnostic imaging and body morphometry. This review provides a comprehensive overview of recent developments and applications of AI in gastric cancer imaging, discussing its role in diagnosis, staging, and postoperative assessment, as well as its limitations.
JOURNAL OF GASTRIC CANCER
(2023)
Article
Medical Informatics
Sora Kang, Chul Park, Jinseok Lee, Dukyong Yoon
Summary: The study developed a machine learning model to predict high-risk bleeding patients in the intensive care unit (ICU) by learning patterns from real clinical data.
HEALTHCARE INFORMATICS RESEARCH
(2022)
Article
Computer Science, Information Systems
Heewon Chung, Jinseok Lee
Summary: One of the challenging issues in classification problems is obtaining enough labeled data for training. Most datasets exhibit class imbalance, which biases the model towards the majority class. To address this issue, semi-supervised learning methods using additional unlabeled data have been proposed. In this study, we propose iterative semi-supervised learning algorithms that correct the labeling of extra unlabeled data based on softmax probabilities. The results show that our algorithms achieve high accuracy comparable to supervised learning. Tested on both balanced and imbalanced unlabeled datasets, our algorithms outperform previous state-of-the-art methods.
CMC-COMPUTERS MATERIALS & CONTINUA
(2022)