Article
Computer Science, Artificial Intelligence
Taha Bugra Celik, Ozgur Ican, Elif Bulut
Summary: Prediction with higher accuracy is important for stock market prediction. However, the black box nature of machine learning techniques and the complexity of the predicted time series limit further improvements in accuracy. Therefore, we propose an eXplainable Artificial Intelligence (XAI) approach to assess prediction reliability and prevent poor decisions.
APPLIED SOFT COMPUTING
(2023)
Article
Engineering, Industrial
Weihong Grace Guo, Vidita Gawade, Bi Zhang, Yuebin Guo
Summary: Explainable Artificial Intelligence is used in this study to improve the understanding of melt pool dynamics in powder bed-based laser fusion. The development of physics-based models and conventional black-box data-driven models to simulate these behaviors proves to be very challenging. A Shapley Additive Explanations (SHAP)-enabled Deep Neural Network-Long Short-Term Memory (DNN-LSTM) model is proposed to integrate process parameter knowledge with process history information using online sensing data, while providing local and global model interpretation and transparency.
CIRP ANNALS-MANUFACTURING TECHNOLOGY
(2023)
Article
Psychology, Multidisciplinary
Mica R. Endsley
Summary: System autonomy and AI are being developed for various applications and will work with humans as human-AI teams (HAT). Situation awareness (SA) is crucial for effective interaction and oversight of these systems. As AI capabilities grow, shared SA between humans and AI becomes more important. Methods for supporting team SA in HAT are discussed, including requirements, mechanisms, displays, and processes. The article also explores the challenges of AI transparency and explainability in supporting SA and mental models. The SA Oriented Design (SAOD) process is described as a systematic methodology for developing transparent AI displays for HAT, with an example in automated driving.
COMPUTERS IN HUMAN BEHAVIOR
(2023)
Article
Environmental Sciences
Abhirup Dikshit, Biswajeet Pradhan
Summary: Accurately predicting natural hazards, especially drought, is challenging. Including climatic variables in data-driven prediction models improves accuracy. Using explainable artificial intelligence models can help understand local interactions during different drought conditions and periods.
SCIENCE OF THE TOTAL ENVIRONMENT
(2021)
Article
Computer Science, Artificial Intelligence
Joao Gabriel Correa Krueger, Alceu de Souza Britto Jr, Jean Paul Barddal
Summary: School dropout is a global socio-economic problem. Predictive models are developed to determine the likelihood of students dropping out. This paper proposes an approach for creating and enriching a dropout prediction dataset using data from 19 schools in Brazil. Experiments achieved high precision, recall, and KS rates when predicting dropout at different year moments. The study also identifies potential reasons for student dropout.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Hardware & Architecture
Priti Shaw, Kaustubh Pachpor, Suresh Sankaranarayanan
Summary: Neonatal sepsis is a significant public health problem, particularly in developing countries, and is the third leading cause of neonatal mortality. Various studies have been conducted on sepsis, vaccine response, and immunity. Machine learning models have been utilized to predict infant mortality based on important features such as age, birth weight, gestational weeks, and APGAR score, but sepsis has not been considered in these predictions. In this study, a deep neural model was deployed to predict infant mortality, taking sepsis into account as a crucial feature. Additionally, explainable AI models like Dalex and Lime were employed to assess the reliability of the deep neural model, providing understandable insights for non-technical personnel such as doctors and practitioners.
COMPUTER SYSTEMS SCIENCE AND ENGINEERING
(2023)
Article
Computer Science, Artificial Intelligence
Oskar Wysocki, Jessica Katharine Davies, Markel Vigo, Anne Caroline Armstrong, Donal Landers, Rebecca Lee, Andre Freitas
Summary: This study presents a pragmatic evaluation framework for explainable Machine Learning (ML) models in clinical decision support. The findings reveal both positive and negative effects of ML explanation models when embedded in the clinical context. However, the study also identifies significant positive effects, such as reducing automation bias and supporting less experienced healthcare professionals in acquiring new domain knowledge.
ARTIFICIAL INTELLIGENCE
(2023)
Article
Computer Science, Information Systems
Harshiv Chandra, Pranav M. Pawar, R. Elakkiya, P. S. Tamizharasan, Raja Muthalagu, Alavikunhu Panthakkan
Summary: Soil fertility refers to the ability of soil in a particular area to provide favorable characteristics for plant growth. This paper discusses the implementation of an explainable AI model that accurately predicts soil fertility using various physiochemical properties and provides user-friendly explanations. The model shows 97.02% accuracy and has applications in improving soil fertility in both the short term and long term.
Article
Biochemical Research Methods
Md Rezaul Karim, Tanhim Islam, Md Shajalal, Oya Beyan, Christoph Lange, Michael Cochez, Dietrich Rebholz-Schuhmann, Stefan Decker
Summary: Artificial intelligence (AI) systems are widely used for solving critical problems in bioinformatics, biomedical informatics, and precision medicine. However, the lack of transparency in complex AI models can be a challenge in understanding their decision-making processes. Explainable AI (XAI) aims to provide transparency and fairness in AI systems, which is particularly important in sensitive areas like healthcare. This paper discusses the importance of explainability in bioinformatics and showcases model-specific and model-agnostic interpretable ML methods that can be customized for bioinformatics research problems. Through case studies, the authors demonstrate how XAI methods can improve transparency and decision fairness in bioinformatics.
BRIEFINGS IN BIOINFORMATICS
(2023)
Article
Chemistry, Physical
Faiza Qayyum, Murad Ali Khan, Do-Hyeun Kim, Hyunseok Ko, Ga-Ae Ryu
Summary: The scientific community is concerned about the transparency and interpretability of machine learning models in materials science. This study uses the TabNet deep learning framework to predict the dielectric constant property of PZT ceramics and analyzes the results using SHAP. The TabNet model outperforms traditional machine learning models, identifying key contributing factors and highlighting their importance.
Review
Biochemistry & Molecular Biology
Thanh Hoa Vo, Ngan Thi Kim Nguyen, Quang Hien Kha, Nguyen Quoc Khanh Le
Summary: Unwanted drug-drug interactions in multi-diseases treatment remain a significant issue. Artificial intelligence (AI) prediction models have potential but concerns about their reliability due to their black-box nature. Building AI models with explainable mechanisms can enhance transparency and promote safety and clarity.
COMPUTATIONAL AND STRUCTURAL BIOTECHNOLOGY JOURNAL
(2022)
Article
Computer Science, Theory & Methods
Elham Khodabandehloo, Daniele Riboni, Abbas Alimohammadi
Summary: The current trend in research is to use sensorized smart-homes and artificial intelligence methods to early detect symptoms of cognitive decline in the elderly. While these tools may provide accurate predictions, they currently have limited support for clinicians in making a diagnosis. This paper proposes a flexible AI system that can recognize early symptoms of cognitive decline in smart-homes and explain the reasons for predictions at a fine-grained level.
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
(2021)
Article
Computer Science, Information Systems
Erzhena Tcydenova, Tae Woo Kim, Changhoon Lee, Jong Hyuk Park
Summary: This paper proposes an adversarial attack detection framework in machine learning-based intrusion detection systems, which detects adversarial attacks by explaining normal data records.
HUMAN-CENTRIC COMPUTING AND INFORMATION SCIENCES
(2021)
Article
Computer Science, Information Systems
Dominik Bork, Syed Juned Ali, Georgi Milenov Dinev
Summary: The Decision Model and Notation (DMN) is a modeling language that precisely specifies business decisions and rules, and is easily understood by business users. However, as models become complex, human cognitive abilities threaten manual maintainability and comprehension. This paper explores the benefits of combining human-driven DMN modeling with Artificial Intelligence for improved analysis and understanding. The authors propose a model-driven approach using DMN models to generate Machine Learning (ML) training data and show how ML models can inform human decision modelers by highlighting feature importance in the original DMN models. An evaluation with real DMN models from an insurance company tests the feasibility and utility of the approach.
BUSINESS & INFORMATION SYSTEMS ENGINEERING
(2023)
Article
Computer Science, Artificial Intelligence
Swarn Avinash Kumar, Moustafa M. Nasralla, Ivan Garcia-Magarino, Harsh Kumar
Summary: This article proposes an expert system that combines machine learning and sentiment analysis to aid small-medium enterprises in determining the best commercialization techniques based on the evolution of the pandemic and customer sentiments.
PEERJ COMPUTER SCIENCE
(2021)
Article
Medicine, General & Internal
Gyeong Hun Kim, Gyoohwan Jung, Jungyo Suh, Juhyun Park, Sung Yong Cho
Summary: The aim of this study was to assess the level of hematuria and presence of clots in RIRS and mPCNL procedures and their impact on surgical outcomes. The newly developed HG system showed excellent intra-observer reliability and correlation with stone density and surgical difficulty.
JOURNAL OF CLINICAL MEDICINE
(2023)
Article
Medicine, General & Internal
Sung Jin Kim, Myungchan Park, Sangjun Yoo
Summary: Hematuria, characterized by red blood cells in the urine, requires immediate investigation for potential urologic cancers, especially in cases of gross hematuria. This study comprehensively reviews various urologic malignancies causing hematuria based on clinical guidelines and published literature.
JOURNAL OF THE KOREAN MEDICAL ASSOCIATION
(2023)
Review
Surgery
Jun Gyo Gwon, Yong-Pil Cho, Youngjin Han, Jungyo Suh, Seung-Kee Min
Summary: Radical nephrectomy with tumor thrombectomy is a challenging surgical technique for advanced renal cell carcinoma. In this study, technical recommendations for tumor thrombectomy were presented, including the use of transesophageal echocardiography and considering cardiopulmonary bypass in certain cases. In addition, the importance of sequential clamping and implementing the Pringle maneuver during IVC cavotomy was highlighted.
VASCULAR SPECIALIST INTERNATIONAL
(2023)
Article
Urology & Nephrology
Sangjun Yoo, Kyung Hee Lee, Parivash Jamrasi, Min Chul Cho, Wook Song, Hyeon Jeong
Summary: In this study, the effects of exercise on the physical function and health quality of life (hQoL) in prostate cancer patients underwent androgen deprivation therapy (ADT) were assessed. The effects of high-intensity interval training (HIIT) and moderate-intensity continuous training (MICT) on these patients were compared. The results showed that the exercise program had a positive effect on the physical function and hQoL of the patients, and HIIT was more effective for improving muscle endurance while MICT was more effective for improving muscle strength.
UROLOGIA INTERNATIONALIS
(2023)
Article
Urology & Nephrology
Jooho Lee, Jung Hoon Lee, Min Soo Choo, Min Chul Cho, Hwancheol Son, Hyeon Jeong, Ji Bong Jeong, Sangjun Yoo
Summary: This study aimed to investigate the practicality of percent body fat (PBF), calculated using bioelectrical impedance analysis (BIA), in predicting benign prostatic hyperplasia/lower urinary tract symptoms. The study found that PBF and appendicular skeletal muscle mass index (ASMI) are useful for predicting BPH/LUTS, and lowering PBF and maintaining adequate ASMI could lower the occurrence of LUTS.
WORLD JOURNAL OF UROLOGY
(2023)
Article
Urology & Nephrology
Bumjin Lim, Sangwook Lee, Taehun Kim, Junhyeok Ock, Cheryn Song, Namkug Kim, Yoon Soo Kyung
Summary: Partial nephrectomy (PN) is a common surgery for small renal masses, aiming to completely remove the mass while preserving renal function. In this study, we used a 3D printing method to create a surgical guide for PN, which accurately indicated the incision line and improved surgical outcome. The guide was easy to handle and showed no complications.
UROLOGIA INTERNATIONALIS
(2023)
Article
Urology & Nephrology
Hyun Sik Yoon, Dae Hyuk Chung, Sung Yong Cho, Min Chul Cho, Jae-Seung Paick, Seung-June Oh
Summary: This study aimed to identify risk factors for salvage procedures (SP) required for refractory adenomatous tissue resistant to morcellation during holmium laser enucleation of the prostate (HoLEP). The results showed that age over 60 and transition zone volume over 32 mL were associated with an increased risk of refractory morcellation. It is necessary to develop more efficient morcellators in the future.
INTERNATIONAL NEUROUROLOGY JOURNAL
(2023)
Article
Urology & Nephrology
Jungyo Suh, Min Soo Choo, Seung-June Oh
Summary: This study evaluated the efficacy and safety of low-power Holmium laser enucleation of the prostate (HoLEP) compared with high-power surgery for benign prostatic hyperplasia (BPH). The results showed that the low-power group had longer operative time but lower delivered energy, faster recovery, and significantly improved surgical outcomes at mid-term follow-up.
INVESTIGATIVE AND CLINICAL UROLOGY
(2023)
Article
Andrology
Sangjun Yoo, Hyeon Jeong, Hwancheol Son, Seung-June Oh, Jae-Seung Paick, Min Chul Cho
Summary: This study examined the effects of preoperative bladder compliance on the long-term functional outcomes after laser prostatectomy. The results showed that laser prostatectomy could lead to further improvement in storage symptoms in patients with decreased bladder compliance.
WORLD JOURNAL OF MENS HEALTH
(2023)
Article
Medicine, General & Internal
Gyoohwan Jung, Seung Min Lee, Sang Won So, Sehwan Kim, Seong Chan Kim, Ohbin Kwon, Hyunjae Song, Min Joo Choi, Sung Yong Cho
Summary: This study examined different patterns of laser fiber degradation and the role of cavitation bubbles in minimizing endoscope damage. The results showed that regardless of laser settings, damages rapidly increased with longer emission time, and the durability of the laser fibers was influenced by the specific settings. The movement of cavitation bubbles contributed to the heat damage around the fiber tips.
JOURNAL OF KOREAN MEDICAL SCIENCE
(2022)