Post-Operative Medium- and Long-Term Endocrine Outcomes in Patients with Non-Functioning Pituitary Adenomas—Machine Learning Analysis
Post-operative endocrine outcomes in patients with non-functioning pituitary adenoma (NFPA) are variable. The aim of this study was to use machine learning (ML) models to better predict medium- and long-term post-operative hypopituitarism in patients with NFPAs. We included data from 383 patients wh...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-05-01
|
| Series: | Cancers |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-6694/15/10/2771 |
| _version_ | 1797600709352357888 |
|---|---|
| author | Ziad Hussein Robert W. Slack Hani J. Marcus Evangelos B. Mazomenos Stephanie E. Baldeweg |
| author_facet | Ziad Hussein Robert W. Slack Hani J. Marcus Evangelos B. Mazomenos Stephanie E. Baldeweg |
| author_sort | Ziad Hussein |
| collection | DOAJ |
| description | Post-operative endocrine outcomes in patients with non-functioning pituitary adenoma (NFPA) are variable. The aim of this study was to use machine learning (ML) models to better predict medium- and long-term post-operative hypopituitarism in patients with NFPAs. We included data from 383 patients who underwent surgery with or without radiotherapy for NFPAs, with a follow-up period between 6 months and 15 years. ML models, including k-nearest neighbour (KNN), support vector machine (SVM), and decision tree models, showed a superior ability to predict panhypopituitarism compared with non-parametric statistical modelling (mean accuracy: 0.89; mean AUC-ROC: 0.79), with SVM achieving the highest performance (mean accuracy: 0.94; mean AUC-ROC: 0.88). Pre-operative endocrine function was the strongest feature for predicting panhypopituitarism within 1 year post-operatively, while endocrine outcomes at 1 year post-operatively supported strong predictions of panhypopituitarism at 5 and 10 years post-operatively. Other features found to contribute to panhypopituitarism prediction were age, volume of tumour, and the use of radiotherapy. In conclusion, our study demonstrates that ML models show potential in predicting post-operative panhypopituitarism in the medium and long term in patients with NFPM. Future work will include incorporating additional, more granular data, including imaging and operative video data, across multiple centres. |
| first_indexed | 2024-03-11T03:51:42Z |
| format | Article |
| id | doaj.art-6ad291d15fd847bc845d75bb13faf187 |
| institution | Directory Open Access Journal |
| issn | 2072-6694 |
| language | English |
| last_indexed | 2024-03-11T03:51:42Z |
| publishDate | 2023-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Cancers |
| spelling | doaj.art-6ad291d15fd847bc845d75bb13faf1872023-11-18T00:48:42ZengMDPI AGCancers2072-66942023-05-011510277110.3390/cancers15102771Post-Operative Medium- and Long-Term Endocrine Outcomes in Patients with Non-Functioning Pituitary Adenomas—Machine Learning AnalysisZiad Hussein0Robert W. Slack1Hani J. Marcus2Evangelos B. Mazomenos3Stephanie E. Baldeweg4Department of Diabetes & Endocrinology, Sheffield Teaching Hospitals NHS Foundation Trust, Sheffield S10 2JF, UKWellcome/EPSRC Centre for Interventional and Surgical Sciences, University College London, London W1W 7TY, UKWellcome/EPSRC Centre for Interventional and Surgical Sciences, University College London, London W1W 7TY, UKWellcome/EPSRC Centre for Interventional and Surgical Sciences, University College London, London W1W 7TY, UKDepartment of Diabetes & Endocrinology, University College London Hospital, London NW1 2BU, UKPost-operative endocrine outcomes in patients with non-functioning pituitary adenoma (NFPA) are variable. The aim of this study was to use machine learning (ML) models to better predict medium- and long-term post-operative hypopituitarism in patients with NFPAs. We included data from 383 patients who underwent surgery with or without radiotherapy for NFPAs, with a follow-up period between 6 months and 15 years. ML models, including k-nearest neighbour (KNN), support vector machine (SVM), and decision tree models, showed a superior ability to predict panhypopituitarism compared with non-parametric statistical modelling (mean accuracy: 0.89; mean AUC-ROC: 0.79), with SVM achieving the highest performance (mean accuracy: 0.94; mean AUC-ROC: 0.88). Pre-operative endocrine function was the strongest feature for predicting panhypopituitarism within 1 year post-operatively, while endocrine outcomes at 1 year post-operatively supported strong predictions of panhypopituitarism at 5 and 10 years post-operatively. Other features found to contribute to panhypopituitarism prediction were age, volume of tumour, and the use of radiotherapy. In conclusion, our study demonstrates that ML models show potential in predicting post-operative panhypopituitarism in the medium and long term in patients with NFPM. Future work will include incorporating additional, more granular data, including imaging and operative video data, across multiple centres.https://www.mdpi.com/2072-6694/15/10/2771non-functioning pituitary adenomahypopituitarismpanhypopituitarismmachine learninglogistic regressionknn |
| spellingShingle | Ziad Hussein Robert W. Slack Hani J. Marcus Evangelos B. Mazomenos Stephanie E. Baldeweg Post-Operative Medium- and Long-Term Endocrine Outcomes in Patients with Non-Functioning Pituitary Adenomas—Machine Learning Analysis Cancers non-functioning pituitary adenoma hypopituitarism panhypopituitarism machine learning logistic regression knn |
| title | Post-Operative Medium- and Long-Term Endocrine Outcomes in Patients with Non-Functioning Pituitary Adenomas—Machine Learning Analysis |
| title_full | Post-Operative Medium- and Long-Term Endocrine Outcomes in Patients with Non-Functioning Pituitary Adenomas—Machine Learning Analysis |
| title_fullStr | Post-Operative Medium- and Long-Term Endocrine Outcomes in Patients with Non-Functioning Pituitary Adenomas—Machine Learning Analysis |
| title_full_unstemmed | Post-Operative Medium- and Long-Term Endocrine Outcomes in Patients with Non-Functioning Pituitary Adenomas—Machine Learning Analysis |
| title_short | Post-Operative Medium- and Long-Term Endocrine Outcomes in Patients with Non-Functioning Pituitary Adenomas—Machine Learning Analysis |
| title_sort | post operative medium and long term endocrine outcomes in patients with non functioning pituitary adenomas machine learning analysis |
| topic | non-functioning pituitary adenoma hypopituitarism panhypopituitarism machine learning logistic regression knn |
| url | https://www.mdpi.com/2072-6694/15/10/2771 |
| work_keys_str_mv | AT ziadhussein postoperativemediumandlongtermendocrineoutcomesinpatientswithnonfunctioningpituitaryadenomasmachinelearninganalysis AT robertwslack postoperativemediumandlongtermendocrineoutcomesinpatientswithnonfunctioningpituitaryadenomasmachinelearninganalysis AT hanijmarcus postoperativemediumandlongtermendocrineoutcomesinpatientswithnonfunctioningpituitaryadenomasmachinelearninganalysis AT evangelosbmazomenos postoperativemediumandlongtermendocrineoutcomesinpatientswithnonfunctioningpituitaryadenomasmachinelearninganalysis AT stephanieebaldeweg postoperativemediumandlongtermendocrineoutcomesinpatientswithnonfunctioningpituitaryadenomasmachinelearninganalysis |