Deep Learning for Prediction of Progression and Recurrence in Nonfunctioning Pituitary Macroadenomas: Combination of Clinical and MRI Features
ObjectivesA subset of non-functioning pituitary macroadenomas (NFMAs) may exhibit early progression/recurrence (P/R) after tumor resection. The purpose of this study was to apply deep learning (DL) algorithms for prediction of P/R in NFMAs.MethodsFrom June 2009 to December 2019, 78 patients diagnose...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2022-04-01
|
| Series: | Frontiers in Oncology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fonc.2022.813806/full |
| _version_ | 1818028573605232640 |
|---|---|
| author | Yan-Jen Chen Yan-Jen Chen Hsun-Ping Hsieh Kuo-Chuan Hung Kuo-Chuan Hung Yun-Ju Shih Sher-Wei Lim Sher-Wei Lim Yu-Ting Kuo Yu-Ting Kuo Jeon-Hor Chen Jeon-Hor Chen Ching-Chung Ko Ching-Chung Ko Ching-Chung Ko |
| author_facet | Yan-Jen Chen Yan-Jen Chen Hsun-Ping Hsieh Kuo-Chuan Hung Kuo-Chuan Hung Yun-Ju Shih Sher-Wei Lim Sher-Wei Lim Yu-Ting Kuo Yu-Ting Kuo Jeon-Hor Chen Jeon-Hor Chen Ching-Chung Ko Ching-Chung Ko Ching-Chung Ko |
| author_sort | Yan-Jen Chen |
| collection | DOAJ |
| description | ObjectivesA subset of non-functioning pituitary macroadenomas (NFMAs) may exhibit early progression/recurrence (P/R) after tumor resection. The purpose of this study was to apply deep learning (DL) algorithms for prediction of P/R in NFMAs.MethodsFrom June 2009 to December 2019, 78 patients diagnosed with pathologically confirmed NFMAs, and who had undergone complete preoperative MRI and postoperative MRI follow-up for more than one year, were included. DL classifiers including multi-layer perceptron (MLP) and convolutional neural network (CNN) were used to build predictive models. Categorical and continuous clinical data were fed into the MLP model, and images of preoperative MRI (T2WI and contrast enhanced T1WI) were analyzed by the CNN model. MLP, CNN and multimodal CNN-MLP architectures were performed to predict P/R in NFMAs.ResultsForty-two (42/78, 53.8%) patients exhibited P/R after surgery. The median follow-up time was 42 months, and the median time to P/R was 25 months. As compared with CNN using MRI (accuracy 83%, precision 87%, and AUC 0.84) or MLP using clinical data (accuracy 73%, precision 73%, and AUC 0.73) alone, the multimodal CNN-MLP model using both clinical and MRI features showed the best performance for prediction of P/R in NFMAs, with accuracy 83%, precision 90%, and AUC 0.85.ConclusionsDL architecture incorporating clinical and MRI features performs well to predict P/R in NFMAs. Pending more studies to support the findings, the results of this study may provide valuable information for NFMAs treatment planning. |
| first_indexed | 2024-12-10T05:05:57Z |
| format | Article |
| id | doaj.art-382d270eaaea4d5392e1d203c144830e |
| institution | Directory Open Access Journal |
| issn | 2234-943X |
| language | English |
| last_indexed | 2024-12-10T05:05:57Z |
| publishDate | 2022-04-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Oncology |
| spelling | doaj.art-382d270eaaea4d5392e1d203c144830e2022-12-22T02:01:14ZengFrontiers Media S.A.Frontiers in Oncology2234-943X2022-04-011210.3389/fonc.2022.813806813806Deep Learning for Prediction of Progression and Recurrence in Nonfunctioning Pituitary Macroadenomas: Combination of Clinical and MRI FeaturesYan-Jen Chen0Yan-Jen Chen1Hsun-Ping Hsieh2Kuo-Chuan Hung3Kuo-Chuan Hung4Yun-Ju Shih5Sher-Wei Lim6Sher-Wei Lim7Yu-Ting Kuo8Yu-Ting Kuo9Jeon-Hor Chen10Jeon-Hor Chen11Ching-Chung Ko12Ching-Chung Ko13Ching-Chung Ko14Department of Electrical Engineering, National Cheng Kung University, Tainan, TaiwanGraduate Institute of Electronics Engineering, National Taiwan University, Taipei, TaiwanDepartment of Electrical Engineering, National Cheng Kung University, Tainan, TaiwanDepartment of Anesthesiology, Chi Mei Medical Center, Tainan, TaiwanDepartment of Hospital and Health Care Administration, College of Recreation and Health Management, Chia Nan University of Pharmacy and Science, Tainan, TaiwanDepartment of Medical Imaging, Chi-Mei Medical Center, Tainan, TaiwanDepartment of Neurosurgery, Chi Mei Medical Center, Tainan, TaiwanDepartment of Nursing, Min-Hwei College of Health Care Management, Tainan, TaiwanDepartment of Medical Imaging, Chi-Mei Medical Center, Tainan, TaiwanDepartment of Medical Imaging, Kaohsiung Medical University Hospital, Kaohsiung, TaiwanDepartment of Radiological Sciences, University of California, Irvine, Irvine, CA, United States0Department of Radiology, E-DA Hospital, I-Shou University, Kaohsiung, TaiwanDepartment of Medical Imaging, Chi-Mei Medical Center, Tainan, Taiwan1Department of Health and Nutrition, Chia Nan University of Pharmacy and Science, Tainan, Taiwan2Institute of Biomedical Sciences, National Sun Yat-Sen University, Kaohsiung, TaiwanObjectivesA subset of non-functioning pituitary macroadenomas (NFMAs) may exhibit early progression/recurrence (P/R) after tumor resection. The purpose of this study was to apply deep learning (DL) algorithms for prediction of P/R in NFMAs.MethodsFrom June 2009 to December 2019, 78 patients diagnosed with pathologically confirmed NFMAs, and who had undergone complete preoperative MRI and postoperative MRI follow-up for more than one year, were included. DL classifiers including multi-layer perceptron (MLP) and convolutional neural network (CNN) were used to build predictive models. Categorical and continuous clinical data were fed into the MLP model, and images of preoperative MRI (T2WI and contrast enhanced T1WI) were analyzed by the CNN model. MLP, CNN and multimodal CNN-MLP architectures were performed to predict P/R in NFMAs.ResultsForty-two (42/78, 53.8%) patients exhibited P/R after surgery. The median follow-up time was 42 months, and the median time to P/R was 25 months. As compared with CNN using MRI (accuracy 83%, precision 87%, and AUC 0.84) or MLP using clinical data (accuracy 73%, precision 73%, and AUC 0.73) alone, the multimodal CNN-MLP model using both clinical and MRI features showed the best performance for prediction of P/R in NFMAs, with accuracy 83%, precision 90%, and AUC 0.85.ConclusionsDL architecture incorporating clinical and MRI features performs well to predict P/R in NFMAs. Pending more studies to support the findings, the results of this study may provide valuable information for NFMAs treatment planning.https://www.frontiersin.org/articles/10.3389/fonc.2022.813806/fulldeep learningpituitarymacroadenomaprogressionrecurrenceMRI |
| spellingShingle | Yan-Jen Chen Yan-Jen Chen Hsun-Ping Hsieh Kuo-Chuan Hung Kuo-Chuan Hung Yun-Ju Shih Sher-Wei Lim Sher-Wei Lim Yu-Ting Kuo Yu-Ting Kuo Jeon-Hor Chen Jeon-Hor Chen Ching-Chung Ko Ching-Chung Ko Ching-Chung Ko Deep Learning for Prediction of Progression and Recurrence in Nonfunctioning Pituitary Macroadenomas: Combination of Clinical and MRI Features Frontiers in Oncology deep learning pituitary macroadenoma progression recurrence MRI |
| title | Deep Learning for Prediction of Progression and Recurrence in Nonfunctioning Pituitary Macroadenomas: Combination of Clinical and MRI Features |
| title_full | Deep Learning for Prediction of Progression and Recurrence in Nonfunctioning Pituitary Macroadenomas: Combination of Clinical and MRI Features |
| title_fullStr | Deep Learning for Prediction of Progression and Recurrence in Nonfunctioning Pituitary Macroadenomas: Combination of Clinical and MRI Features |
| title_full_unstemmed | Deep Learning for Prediction of Progression and Recurrence in Nonfunctioning Pituitary Macroadenomas: Combination of Clinical and MRI Features |
| title_short | Deep Learning for Prediction of Progression and Recurrence in Nonfunctioning Pituitary Macroadenomas: Combination of Clinical and MRI Features |
| title_sort | deep learning for prediction of progression and recurrence in nonfunctioning pituitary macroadenomas combination of clinical and mri features |
| topic | deep learning pituitary macroadenoma progression recurrence MRI |
| url | https://www.frontiersin.org/articles/10.3389/fonc.2022.813806/full |
| work_keys_str_mv | AT yanjenchen deeplearningforpredictionofprogressionandrecurrenceinnonfunctioningpituitarymacroadenomascombinationofclinicalandmrifeatures AT yanjenchen deeplearningforpredictionofprogressionandrecurrenceinnonfunctioningpituitarymacroadenomascombinationofclinicalandmrifeatures AT hsunpinghsieh deeplearningforpredictionofprogressionandrecurrenceinnonfunctioningpituitarymacroadenomascombinationofclinicalandmrifeatures AT kuochuanhung deeplearningforpredictionofprogressionandrecurrenceinnonfunctioningpituitarymacroadenomascombinationofclinicalandmrifeatures AT kuochuanhung deeplearningforpredictionofprogressionandrecurrenceinnonfunctioningpituitarymacroadenomascombinationofclinicalandmrifeatures AT yunjushih deeplearningforpredictionofprogressionandrecurrenceinnonfunctioningpituitarymacroadenomascombinationofclinicalandmrifeatures AT sherweilim deeplearningforpredictionofprogressionandrecurrenceinnonfunctioningpituitarymacroadenomascombinationofclinicalandmrifeatures AT sherweilim deeplearningforpredictionofprogressionandrecurrenceinnonfunctioningpituitarymacroadenomascombinationofclinicalandmrifeatures AT yutingkuo deeplearningforpredictionofprogressionandrecurrenceinnonfunctioningpituitarymacroadenomascombinationofclinicalandmrifeatures AT yutingkuo deeplearningforpredictionofprogressionandrecurrenceinnonfunctioningpituitarymacroadenomascombinationofclinicalandmrifeatures AT jeonhorchen deeplearningforpredictionofprogressionandrecurrenceinnonfunctioningpituitarymacroadenomascombinationofclinicalandmrifeatures AT jeonhorchen deeplearningforpredictionofprogressionandrecurrenceinnonfunctioningpituitarymacroadenomascombinationofclinicalandmrifeatures AT chingchungko deeplearningforpredictionofprogressionandrecurrenceinnonfunctioningpituitarymacroadenomascombinationofclinicalandmrifeatures AT chingchungko deeplearningforpredictionofprogressionandrecurrenceinnonfunctioningpituitarymacroadenomascombinationofclinicalandmrifeatures AT chingchungko deeplearningforpredictionofprogressionandrecurrenceinnonfunctioningpituitarymacroadenomascombinationofclinicalandmrifeatures |