A Personalized Medical Decision Support System Based on Explainable Machine Learning Algorithms and ECC Features: Data from the Real World

Gu, Dongxiao; Zhao, Wang; Xie, Yi; Wang, Xiaoyu; Su, Kaixiang; Золотарев, Олег

doi:10.3390/diagnostics11091677

Cited by 12 publications

(4 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This reduces the complexity of a medical information system, making it easier to maintain and upgrade [ 55 ]. At the same time, as the models based on deep neural networks have not been able to effectively raise the interpretation to the understanding of the end user [ 56 , 57 ], too-complex models will hinder doctors’ trust in CAD systems [ 12 , 13 ]. In our work, the integration of relatively simple models (such as tree-based models) ensures the system performance while taking into account doctors’ understanding of the model decision-making process.…”

Section: Resultsmentioning

confidence: 99%

An Adaptive Deep Ensemble Learning Method for Dynamic Evolving Diagnostic Task Scenarios

et al. 2021

Diagnostics

Self Cite

View full text Add to dashboard Cite

Increasingly, machine learning methods have been applied to aid in diagnosis with good results. However, some complex models can confuse physicians because they are difficult to understand, while data differences across diagnostic tasks and institutions can cause model performance fluctuations. To address this challenge, we combined the Deep Ensemble Model (DEM) and tree-structured Parzen Estimator (TPE) and proposed an adaptive deep ensemble learning method (TPE-DEM) for dynamic evolving diagnostic task scenarios. Different from previous research that focuses on achieving better performance with a fixed structure model, our proposed model uses TPE to efficiently aggregate simple models more easily understood by physicians and require less training data. In addition, our proposed model can choose the optimal number of layers for the model and the type and number of basic learners to achieve the best performance in different diagnostic task scenarios based on the data distribution and characteristics of the current diagnostic task. We tested our model on one dataset constructed with a partner hospital and five UCI public datasets with different characteristics and volumes based on various diagnostic tasks. Our performance evaluation results show that our proposed model outperforms other baseline models on different datasets. Our study provides a novel approach for simple and understandable machine learning models in tasks with variable datasets and feature sets, and the findings have important implications for the application of machine learning models in computer-aided diagnosis.

show abstract

Section: Resultsmentioning

confidence: 99%

An Adaptive Deep Ensemble Learning Method for Dynamic Evolving Diagnostic Task Scenarios

et al. 2021

Diagnostics

Self Cite

View full text Add to dashboard Cite

show abstract

“…In recent years, with the increase of women's life and work pressure, the incidence of breast cancer has also been rising, and it is expanding toward a younger group [ 14 , 15 ]. At present, radical surgery is the preferred solution for the treatment of breast cancer [ 16 , 17 ].…”

Section: Discussionmentioning

confidence: 99%

Evaluation of Serum Fructosamine as Diagnostic Marker of Postoperative Recurrence in the Patients with Breast Cancer

Zhao

Xie

et al. 2023

BioMed Research International

View full text Add to dashboard Cite

Objectives. A series of laboratory parameters were screened to identify the proper serum markers that could be used to predict breast cancer recurrence at an early stage. Methods. A case-control retrospective study on 224 patients without postoperative recurrence and 43 patients with postoperative recurrence of breast cancer was performed. The edgeR software package was used to identify the test indicators expressed differently between the two groups. Univariate analysis was used to screen for diagnostic marker that could predict postoperative recurrence of breast cancer. In addition, the differential test indicators at different time points from surgery to recurrence were collected in patients with postoperative recurrence of breast cancer as a verification database. Results. We screened out three test indicators (TBA, GSP, and URBC) for differential expression, which were all expressed downregulated in the postoperative recurrence group of breast cancer. Univariate analysis suggested that only the difference in GSP levels between the two groups was statistically significant ( P = 0.001 ). ROC curve analysis showed that the area under the curve of GSP was 0.662, while the area under the curve of GSP+AFP+CEA+CA125+CA153+age was increased to 0.828. In addition, serum GSP levels were significantly reduced after recurrence compared with before recurrence in breast cancer patients ( P < 0.01 ). Conclusions. In summary, GSP could be used for early diagnosis of breast cancer recurrence after surgery, and the predicted value of combining GSP, tumor markers, and age was better than that of individual indicators.

show abstract

“…A continuous increase in demands for health services, whether due to overpopulation, increasing aging, or the emergence of a pandemic, results in an ever-growing need to support health professionals in making decisions [1] and choosing treatment plans according to personal characteristics and the health status of individuals [2]. The ultimate goals of all health systems are to ensure high levels of care and contribute to quality of life [3,4].…”

Section: Introductionmentioning

confidence: 99%

Extracting Interpretable Knowledge from the Remote Monitoring of COVID-19 Patients

Tziomaka,

Kallipolitis,

Menychtas

et al. 2024

MAKE

View full text Add to dashboard Cite

Apart from providing user-friendly applications that support digitized healthcare routines, the use of wearable devices has proven to increase the independence of patients in a healthcare setting. By applying machine learning techniques to real health-related data, important conclusions can be drawn for unsolved issues related to disease prognosis. In this paper, various machine learning techniques are examined and analyzed for the provision of personalized care to COVID-19 patients with mild symptoms based on individual characteristics and the comorbidities they have, while the connection between the stimuli and predictive results are utilized for the evaluation of the system’s transparency. The results, jointly analyzing wearable and electronic health record data for the prediction of a daily dyspnea grade and the duration of fever, are promising in terms of evaluation metrics even in a specified stratum of patients. The interpretability scheme provides useful insight concerning factors that greatly influenced the results. Moreover, it is demonstrated that the use of wearable devices for remote monitoring through cloud platforms is feasible while providing awareness of a patient’s condition, leading to the early detection of undesired changes and reduced visits for patient screening.

show abstract

A Personalized Medical Decision Support System Based on Explainable Machine Learning Algorithms and ECC Features: Data from the Real World

Cited by 12 publications

References 41 publications

An Adaptive Deep Ensemble Learning Method for Dynamic Evolving Diagnostic Task Scenarios

An Adaptive Deep Ensemble Learning Method for Dynamic Evolving Diagnostic Task Scenarios

Evaluation of Serum Fructosamine as Diagnostic Marker of Postoperative Recurrence in the Patients with Breast Cancer

Extracting Interpretable Knowledge from the Remote Monitoring of COVID-19 Patients

Contact Info

Product

Resources

About