2020
DOI: 10.2196/22421
|View full text |Cite
|
Sign up to set email alerts
|

Integrating a Machine Learning System Into Clinical Workflows: Qualitative Study

Abstract: Background Machine learning models have the potential to improve diagnostic accuracy and management of acute conditions. Despite growing efforts to evaluate and validate such models, little is known about how to best translate and implement these products as part of routine clinical care. Objective This study aims to explore the factors influencing the integration of a machine learning sepsis early warning system (Sepsis Watch) into clinical workflows. … Show more

Help me understand this report
View preprint versions

Search citation statements

Order By: Relevance

Paper Sections

Select...
4
1

Citation Types

1
78
1
1

Year Published

2021
2021
2024
2024

Publication Types

Select...
7
1

Relationship

0
8

Authors

Journals

citations
Cited by 95 publications
(81 citation statements)
references
References 47 publications
1
78
1
1
Order By: Relevance
“…Moreover, different medical disciplines were considered in the interviews (eg, radiology, pathology, and internal medicine) to allow for different perspectives on medical diagnostic processes (eg, interpretation of medical scans, pathology slides, and electrocardiograms) and obtain more generalizable results [45]. The resulting number of interviews is comparable with those of other qualitative studies in IS health care research [31,34,46,47]. With regard to data analysis, we followed a structured and reproducible approach to evaluate the qualitative data [36,42].…”
Section: Overviewmentioning
confidence: 99%
See 2 more Smart Citations
“…Moreover, different medical disciplines were considered in the interviews (eg, radiology, pathology, and internal medicine) to allow for different perspectives on medical diagnostic processes (eg, interpretation of medical scans, pathology slides, and electrocardiograms) and obtain more generalizable results [45]. The resulting number of interviews is comparable with those of other qualitative studies in IS health care research [31,34,46,47]. With regard to data analysis, we followed a structured and reproducible approach to evaluate the qualitative data [36,42].…”
Section: Overviewmentioning
confidence: 99%
“…Consistent with the call of Davison and Martinson [ 29 ] for more context-specific research, studies regarding the adoption of ML systems in clinics must, therefore, reflect on both, the specific characteristics of ML systems and clinics. Such context-specific research on the organizational adoption of ML systems in clinics is becoming more prevalent in recent times [ 10 , 30 , 31 ]. Thematically, researchers mainly investigate the individual acceptance of physicians [ 19 , 31 ] and the technical specifics of ML systems, such as their lack of transparency [ 32 , 33 ].…”
Section: Introductionmentioning
confidence: 99%
See 1 more Smart Citation
“…Various models are used to identify drugs associated with the risk of DILI at the preclinical stage [28]. Machine learning models have demonstrated strong predictive power and retained a simple form for communication with researchers [29][30][31][32][33][34][35][36][37][38][39]. XGBoost is a boosting ensemble machine learning algorithm that integrates a few classification and regression trees models to form a strong classifier [40,41].…”
Section: Introductionmentioning
confidence: 99%
“…However, the accuracy of such algorithms can be highly impacted by the complex workflows adopted to develop and generalize such ML algorithms [ 24 , 25 ]. High heterogeneity is expected, as ML problems are usually regarded as black boxes, and the consideration of all possible risk factors and transformation is tremendously difficult [ 26 , 27 ]. Moreover, there are no clear guidelines on how to develop ML approaches for medical studies.…”
Section: Introductionmentioning
confidence: 99%