Customized and Automated Machine Learning-Based Models for Diabetes Type 2 Classification

Mohsen, Farida; Biswas, Md. Rafiul; Ali, Hazrat; Alam, Tanvir; Househ, Mowafa; Shah, Zubair

doi:10.3233/shti220779

Prediction of Early Mortality in Patients Undergoing Right Hemicolectomy Using Machine Learning Algorithms

Kement,

Alkan,

Irmak

et al. 2024

Preprint

0

View full text Add to dashboard Cite

Aim:This study aims to determine whether early mortality in patients undergoing right hemicolectomy can be predicted using artificial intelligence (machine learning) algorithms. Method:The study included all cases of right hemicolectomy or extended right hemicolectomy performed in our clinic between January 2019 and December 2023. Data were collected retrospectively from a prospectively maintained database. Patients undergoing surgeries other than right hemicolectomy were excluded. A database was created using basic clinical data and processed in the Google Colab environment using TensorFlow, Keras, Pandas, Numpy, and Scikit-learn libraries. The TensorFlow. Keras Sequential model was used with “Dense” layers, and the “Adam” optimizer was chosen for optimization. Eighty percent of the database was used for training, and 20% was used for testing. Results:The database included 410 patients, of whom 258 (62.9%) were male, and 152 (37.1%) were female, with a mean age of 63.5±14.2 years. Early mortality occurred in 39 (9.5%) patients. Data from 307 patients were used for machine learning training, and data from 103 patients were used for testing. The machine-learning model predicted early mortality with an accuracy range of 91% to 95% using basic clinical parameters.. Conclusion:This study demonstrates that early mortality in patients undergoing right hemicolectomy can be predicted with high accuracy using machine learning algorithms. The above 90% accuracy rate achieved using basic clinical parameters indicates the potential of this algorithm as a supportive tool in clinical decision-making processes.

show abstract

Clinical performance of automated machine learning: A systematic review

Thirunavukarasu,

Elangovan,

Gutierrez

et al. 2024

Ann Acad Med Singap

0

View full text Add to dashboard Cite

Introduction: Automated machine learning (autoML) removes technical and technological barriers to building artificial intelligence models. We aimed to summarise the clinical applications of autoML, assess the capabilities of utilised platforms, evaluate the quality of the evidence trialling autoML, and gauge the performance of autoML platforms relative to conventionally developed models, as well as each other. Method: This review adhered to a prospectively registered protocol (PROSPERO identifier CRD42022344427). The Cochrane Library, Embase, MEDLINE and Scopus were searched from inception to 11 July 2022. Two researchers screened abstracts and full texts, extracted data and conducted quality assessment. Disagreement was resolved through discussion and as if required, arbitration by a third researcher. Results: There were 26 distinct autoML platforms featured in 82 studies. Brain and lung disease were the most common fields of study of 22 specialties. AutoML exhibited variable performance: area under the receiver operator characteristic curve (AUCROC) 0.35–1.00, F1-score 0.16–0.99, area under the precision-recall curve (AUPRC) 0.51–1.00. AutoML exhibited the highest AUCROC in 75.6% trials; the highest F1-score in 42.3% trials; and the highest AUPRC in 83.3% trials. In autoML platform comparisons, AutoPrognosis and Amazon Rekognition performed strongest with unstructured and structured data, respectively. Quality of reporting was poor, with a median DECIDE-AI score of 14 of 27. Conclusion: A myriad of autoML platforms have been applied in a variety of clinical contexts. The performance of autoML compares well to bespoke computational and clinical benchmarks. Further work is required to improve the quality of validation studies. AutoML may facilitate a transition to data-centric development, and integration with large language models may enable AI to build itself to fulfil user-defined goals.

show abstract

Clinical performance of automated machine learning: A systematic review

Thirunavukarasu,

Elangovan,

Gutierrez

et al. 2024

Ann Acad Med Singap

1

0

View full text Add to dashboard Cite

Introduction: Automated machine learning (autoML) removes technical and technological barriers to building artificial intelligence models. We aimed to summarise the clinical applications of autoML, assess the capabilities of utilised platforms, evaluate the quality of the evidence trialling autoML, and gauge the performance of autoML platforms relative to conventionally developed models, as well as each other. Method: This review adhered to a prospectively registered protocol (PROSPERO identifier CRD42022344427). The Cochrane Library, Embase, MEDLINE and Scopus were searched from inception to 11 July 2022. Two researchers screened abstracts and full texts, extracted data and conducted quality assessment. Disagreement was resolved through discussion and as if required, arbitration by a third researcher. Results: There were 26 distinct autoML platforms featured in 82 studies. Brain and lung disease were the most common fields of study of 22 specialties. AutoML exhibited variable performance: area under the receiver operator characteristic curve (AUCROC) 0.35–1.00, F1-score 0.16–0.99, area under the precision-recall curve (AUPRC) 0.51–1.00. AutoML exhibited the highest AUCROC in 75.6% trials; the highest F1-score in 42.3% trials; and the highest AUPRC in 83.3% trials. In autoML platform comparisons, AutoPrognosis and Amazon Rekognition performed strongest with unstructured and structured data, respectively. Quality of reporting was poor, with a median DECIDE-AI score of 14 of 27. Conclusion: A myriad of autoML platforms have been applied in a variety of clinical contexts. The performance of autoML compares well to bespoke computational and clinical benchmarks. Further work is required to improve the quality of validation studies. AutoML may facilitate a transition to data-centric development, and integration with large language models may enable AI to build itself to fulfil user-defined goals.

show abstract

Customized and Automated Machine Learning-Based Models for Diabetes Type 2 Classification

Cited by 3 publications

References 5 publications

Prediction of Early Mortality in Patients Undergoing Right Hemicolectomy Using Machine Learning Algorithms

Prediction of Early Mortality in Patients Undergoing Right Hemicolectomy Using Machine Learning Algorithms

Clinical performance of automated machine learning: A systematic review

Clinical performance of automated machine learning: A systematic review

Contact Info

Product

Resources

About