1
|
Vallevik VB, Babic A, Marshall SE, Elvatun S, Brøgger HMB, Alagaratnam S, Edwin B, Veeraragavan NR, Befring AK, Nygård JF. Can I trust my fake data - A comprehensive quality assessment framework for synthetic tabular data in healthcare. Int J Med Inform 2024; 185:105413. [PMID: 38493547 DOI: 10.1016/j.ijmedinf.2024.105413] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/15/2023] [Revised: 02/17/2024] [Accepted: 03/11/2024] [Indexed: 03/19/2024]
Abstract
BACKGROUND Ensuring safe adoption of AI tools in healthcare hinges on access to sufficient data for training, testing and validation. Synthetic data has been suggested in response to privacy concerns and regulatory requirements and can be created by training a generator on real data to produce a dataset with similar statistical properties. Competing metrics with differing taxonomies for quality evaluation have been proposed, resulting in a complex landscape. Optimising quality entails balancing considerations that make the data fit for use, yet relevant dimensions are left out of existing frameworks. METHOD We performed a comprehensive literature review on the use of quality evaluation metrics on synthetic data within the scope of synthetic tabular healthcare data using deep generative methods. Based on this and the collective team experiences, we developed a conceptual framework for quality assurance. The applicability was benchmarked against a practical case from the Dutch National Cancer Registry. CONCLUSION We present a conceptual framework for quality assuranceof synthetic data for AI applications in healthcare that aligns diverging taxonomies, expands on common quality dimensions to include the dimensions of Fairness and Carbon footprint, and proposes stages necessary to support real-life applications. Building trust in synthetic data by increasing transparency and reducing the safety risk will accelerate the development and uptake of trustworthy AI tools for the benefit of patients. DISCUSSION Despite the growing emphasis on algorithmic fairness and carbon footprint, these metrics were scarce in the literature review. The overwhelming focus was on statistical similarity using distance metrics while sequential logic detection was scarce. A consensus-backed framework that includes all relevant quality dimensions can provide assurance for safe and responsible real-life applications of synthetic data. As the choice of appropriate metrics are highly context dependent, further research is needed on validation studies to guide metric choices and support the development of technical standards.
Collapse
Affiliation(s)
- Vibeke Binz Vallevik
- University of Oslo, Boks 1072 Blindern, NO-0316 Oslo, Norway; DNV AS, Veritasveien 1, 1322 Høvik, Norway.
| | | | | | - Severin Elvatun
- Cancer Registry of Norway, Ullernchausseen 64, 0379 Oslo, Norway
| | - Helga M B Brøgger
- DNV AS, Veritasveien 1, 1322 Høvik, Norway; Oslo University Hospital, Sognsvannsveien 20, 0372 Oslo, Norway
| | | | - Bjørn Edwin
- University of Oslo, Boks 1072 Blindern, NO-0316 Oslo, Norway; The Intervention Centre and Department of HPB Surgery, Oslo University Hospital and Institute of Clinical Medicine, Faculty of Medicine, University of Oslo, Oslo, Norway
| | | | | | - Jan F Nygård
- Cancer Registry of Norway, Ullernchausseen 64, 0379 Oslo, Norway; UiT - The Arctic University of Norway, Tromsø, Norway
| |
Collapse
|
2
|
Ghosh SK, Khandoker AH. Investigation on explainable machine learning models to predict chronic kidney diseases. Sci Rep 2024; 14:3687. [PMID: 38355876 PMCID: PMC10866953 DOI: 10.1038/s41598-024-54375-4] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/11/2023] [Accepted: 02/12/2024] [Indexed: 02/16/2024] Open
Abstract
Chronic kidney disease (CKD) is a major worldwide health problem, affecting a large proportion of the world's population and leading to higher morbidity and death rates. The early stages of CKD sometimes present without visible symptoms, causing patients to be unaware. Early detection and treatments are critical in reducing complications and improving the overall quality of life for people afflicted. In this work, we investigate the use of an explainable artificial intelligence (XAI)-based strategy, leveraging clinical characteristics, to predict CKD. This study collected clinical data from 491 patients, comprising 56 with CKD and 435 without CKD, encompassing clinical, laboratory, and demographic variables. To develop the predictive model, five machine learning (ML) methods, namely logistic regression (LR), random forest (RF), decision tree (DT), Naïve Bayes (NB), and extreme gradient boosting (XGBoost), were employed. The optimal model was selected based on accuracy and area under the curve (AUC). Additionally, the SHAP (SHapley Additive exPlanations) and LIME (Local Interpretable Model-agnostic Explanations) algorithms were utilized to demonstrate the influence of the features on the optimal model. Among the five models developed, the XGBoost model achieved the best performance with an AUC of 0.9689 and an accuracy of 93.29%. The analysis of feature importance revealed that creatinine, glycosylated hemoglobin type A1C (HgbA1C), and age were the three most influential features in the XGBoost model. The SHAP force analysis further illustrated the model's visualization of individualized CKD predictions. For further insights into individual predictions, we also utilized the LIME algorithm. This study presents an interpretable ML-based approach for the early prediction of CKD. The SHAP and LIME methods enhance the interpretability of ML models and help clinicians better understand the rationale behind the predicted outcomes more effectively.
Collapse
Affiliation(s)
- Samit Kumar Ghosh
- Department of Biomedical Engineering & Biotechnology, Khalifa University, Abu Dhabi, United Arab Emirates.
| | - Ahsan H Khandoker
- Department of Biomedical Engineering & Biotechnology, Khalifa University, Abu Dhabi, United Arab Emirates
| |
Collapse
|
3
|
Chadaga K, Prabhu S, Sampathila N, Chadaga R, Umakanth S, Bhat D, G S SK. Explainable artificial intelligence approaches for COVID-19 prognosis prediction using clinical markers. Sci Rep 2024; 14:1783. [PMID: 38245638 PMCID: PMC10799946 DOI: 10.1038/s41598-024-52428-2] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/23/2023] [Accepted: 01/18/2024] [Indexed: 01/22/2024] Open
Abstract
The COVID-19 influenza emerged and proved to be fatal, causing millions of deaths worldwide. Vaccines were eventually discovered, effectively preventing the severe symptoms caused by the disease. However, some of the population (elderly and patients with comorbidities) are still vulnerable to severe symptoms such as breathlessness and chest pain. Identifying these patients in advance is imperative to prevent a bad prognosis. Hence, machine learning and deep learning algorithms have been used for early COVID-19 severity prediction using clinical and laboratory markers. The COVID-19 data was collected from two Manipal hospitals after obtaining ethical clearance. Multiple nature-inspired feature selection algorithms are used to choose the most crucial markers. A maximum testing accuracy of 95% was achieved by the classifiers. The predictions obtained by the classifiers have been demystified using five explainable artificial intelligence techniques (XAI). According to XAI, the most important markers are c-reactive protein, basophils, lymphocytes, albumin, D-Dimer and neutrophils. The models could be deployed in various healthcare facilities to predict COVID-19 severity in advance so that appropriate treatments could be provided to mitigate a severe prognosis. The computer aided diagnostic method can also aid the healthcare professionals and ease the burden on already suffering healthcare infrastructure.
Collapse
Affiliation(s)
- Krishnaraj Chadaga
- Department of Computer Science and Engineering, Manipal Institute of Technology, Manipal Academy of Higher Education, Manipal, Karnataka, India.
| | - Srikanth Prabhu
- Department of Computer Science and Engineering, Manipal Institute of Technology, Manipal Academy of Higher Education, Manipal, Karnataka, India.
| | - Niranjana Sampathila
- Department of Biomedical Engineering, Manipal Institute of Technology, Manipal Academy of Higher Education, Manipal, Karnataka, India.
| | - Rajagopala Chadaga
- Department of Mechanical and Industrial Engineering, Manipal Institute of Technology, Manipal Academy of Higher Education, Manipal, Karnataka, India
| | - Shashikiran Umakanth
- Department of Medicine, Dr. TMA Hospital, Manipal Academy of Higher Education, Manipal, Karnataka, India
| | - Devadas Bhat
- Department of Biomedical Engineering, Manipal Institute of Technology, Manipal Academy of Higher Education, Manipal, Karnataka, India
| | - Shashi Kumar G S
- Department of Electronics and Communication Engineering, Manipal Institute of Technology, Manipal Academy of Higher Education, Manipal, Karnataka, India
| |
Collapse
|
4
|
Nagendran M, Festor P, Komorowski M, Gordon AC, Faisal AA. Quantifying the impact of AI recommendations with explanations on prescription decision making. NPJ Digit Med 2023; 6:206. [PMID: 37935953 PMCID: PMC10630476 DOI: 10.1038/s41746-023-00955-z] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/23/2023] [Accepted: 10/27/2023] [Indexed: 11/09/2023] Open
Abstract
The influence of AI recommendations on physician behaviour remains poorly characterised. We assess how clinicians' decisions may be influenced by additional information more broadly, and how this influence can be modified by either the source of the information (human peers or AI) and the presence or absence of an AI explanation (XAI, here using simple feature importance). We used a modified between-subjects design where intensive care doctors (N = 86) were presented on a computer for each of 16 trials with a patient case and prompted to prescribe continuous values for two drugs. We used a multi-factorial experimental design with four arms, where each clinician experienced all four arms on different subsets of our 24 patients. The four arms were (i) baseline (control), (ii) peer human clinician scenario showing what doses had been prescribed by other doctors, (iii) AI suggestion and (iv) XAI suggestion. We found that additional information (peer, AI or XAI) had a strong influence on prescriptions (significantly for AI, not so for peers) but simple XAI did not have higher influence than AI alone. There was no correlation between attitudes to AI or clinical experience on the AI-supported decisions and nor was there correlation between what doctors self-reported about how useful they found the XAI and whether the XAI actually influenced their prescriptions. Our findings suggest that the marginal impact of simple XAI was low in this setting and we also cast doubt on the utility of self-reports as a valid metric for assessing XAI in clinical experts.
Collapse
Affiliation(s)
- Myura Nagendran
- UKRI Centre for Doctoral Training in AI for Healthcare, Imperial College London, London, UK
- Division of Anaesthetics, Pain Medicine, and Intensive Care, Imperial College London, London, UK
- Brain and Behaviour Lab, Imperial College London, London, UK
| | - Paul Festor
- UKRI Centre for Doctoral Training in AI for Healthcare, Imperial College London, London, UK
- Brain and Behaviour Lab, Imperial College London, London, UK
- Department of Computing, Imperial College London, London, UK
| | - Matthieu Komorowski
- Division of Anaesthetics, Pain Medicine, and Intensive Care, Imperial College London, London, UK
| | - Anthony C Gordon
- Division of Anaesthetics, Pain Medicine, and Intensive Care, Imperial College London, London, UK
| | - Aldo A Faisal
- UKRI Centre for Doctoral Training in AI for Healthcare, Imperial College London, London, UK.
- Brain and Behaviour Lab, Imperial College London, London, UK.
- Department of Computing, Imperial College London, London, UK.
- Institute of Artificial & Human Intelligence, University of Bayreuth, Bayreuth, Germany.
| |
Collapse
|
5
|
Gao YC, Yao N, Chen X, Yu L, Zhang R, Zhang Q. Data-Driven Insight into the Reductive Stability of Ion-Solvent Complexes in Lithium Battery Electrolytes. J Am Chem Soc 2023; 145:23764-23770. [PMID: 37703183 DOI: 10.1021/jacs.3c08346] [Citation(s) in RCA: 7] [Impact Index Per Article: 7.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 09/15/2023]
Abstract
Lithium (Li) metal batteries (LMBs) are regarded as one of the most promising energy storage systems due to their ultrahigh theoretical energy density. However, the high reactivity of the Li anodes leads to the decomposition of the electrolytes, presenting a huge impediment to the practical application of LMBs. The routine trial-and-error methods are inefficient in designing highly stable solvent molecules for the Li metal anode. Herein, a data-driven approach is proposed to probe the origin of the reductive stability of solvents and accelerate the molecular design for advanced electrolytes. A large database of potential solvent molecules is first constructed using a graph theory-based algorithm and then comprehensively investigated by both first-principles calculations and machine learning (ML) methods. The reductive stability of 99% of the electrolytes decreases under the dominance of ion-solvent complexes, according to the analysis of the lowest unoccupied molecular orbital (LUMO). The LUMO energy level is related to the binding energy, bond length, and orbital ratio factors. An interpretable ML method based on Shapley additive explanations identifies the dipole moment and molecular radius as the most critical descriptors affecting the reductive stability of coordinated solvents. This work not only affords fruitful data-driven insight into the ion-solvent chemistry but also unveils the critical molecular descriptors in regulating the solvent's reductive stability, which accelerates the rational design of advanced electrolyte molecules for next-generation Li batteries.
Collapse
Affiliation(s)
- Yu-Chen Gao
- Beijing Key Laboratory of Green Chemical Reaction Engineering and Technology, Department of Chemical Engineering, Tsinghua University, Beijing 100084, China
| | - Nan Yao
- Beijing Key Laboratory of Green Chemical Reaction Engineering and Technology, Department of Chemical Engineering, Tsinghua University, Beijing 100084, China
| | - Xiang Chen
- Beijing Key Laboratory of Green Chemical Reaction Engineering and Technology, Department of Chemical Engineering, Tsinghua University, Beijing 100084, China
| | - Legeng Yu
- Beijing Key Laboratory of Green Chemical Reaction Engineering and Technology, Department of Chemical Engineering, Tsinghua University, Beijing 100084, China
| | - Rui Zhang
- Beijing Huairou Laboratory, Beijing 101400, China
| | - Qiang Zhang
- Beijing Key Laboratory of Green Chemical Reaction Engineering and Technology, Department of Chemical Engineering, Tsinghua University, Beijing 100084, China
| |
Collapse
|