A Few-Shot Learning Approach Assists in the Prognosis Prediction of Magnetic Resonance-Guided Focused Ultrasound for the Local Control of Bone Metastatic Lesions
Magnetic resonance-guided focused ultrasound surgery (MRgFUS) constitutes a noninvasive treatment strategy to ablate deep-seated bone metastases. However, limited evidence suggests that, although cytokines are influenced by thermal necrosis, there is still no cytokine threshold for clinical response...
| Main Authors: | , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-01-01
|
| Series: | Cancers |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-6694/14/2/445 |
| _version_ | 1797495238231588864 |
|---|---|
| author | Fang-Chi Hsu Hsin-Lun Lee Yin-Ju Chen Yao-An Shen Yi-Chieh Tsai Meng-Huang Wu Chia-Chun Kuo Long-Sheng Lu Shauh-Der Yeh Wen-Sheng Huang Chia-Ning Shen Jeng-Fong Chiou |
| author_facet | Fang-Chi Hsu Hsin-Lun Lee Yin-Ju Chen Yao-An Shen Yi-Chieh Tsai Meng-Huang Wu Chia-Chun Kuo Long-Sheng Lu Shauh-Der Yeh Wen-Sheng Huang Chia-Ning Shen Jeng-Fong Chiou |
| author_sort | Fang-Chi Hsu |
| collection | DOAJ |
| description | Magnetic resonance-guided focused ultrasound surgery (MRgFUS) constitutes a noninvasive treatment strategy to ablate deep-seated bone metastases. However, limited evidence suggests that, although cytokines are influenced by thermal necrosis, there is still no cytokine threshold for clinical responses. A prediction model to approximate the postablation immune status on the basis of circulating cytokine activation is thus needed. IL-6 and IP-10, which are proinflammatory cytokines, decreased significantly during the acute phase. Wound-healing cytokines such as VEGF and PDGF increased after ablation, but the increase was not statistically significant. In this phase, IL-6, IL-13, IP-10, and eotaxin expression levels diminished the ongoing inflammatory progression in the treated sites. These cytokine changes also correlated with the response rate of primary tumor control after acute periods. The few-shot learning algorithm was applied to test the correlation between cytokine levels and local control (<i>p</i> = 0.036). The best-fitted model included IL-6, IL-13, IP-10, and eotaxin as cytokine parameters from the few-shot selection, and had an accuracy of 85.2%, sensitivity of 88.6%, and AUC of 0.95. The acceptable usage of this model may help predict the acute-phase prognosis of a patient with painful bone metastasis who underwent local MRgFUS. The application of machine learning in bone metastasis is equivalent or better than the current logistic regression. |
| first_indexed | 2024-03-10T01:45:59Z |
| format | Article |
| id | doaj.art-283959d772e2423b8b9b4163cdd957cc |
| institution | Directory Open Access Journal |
| issn | 2072-6694 |
| language | English |
| last_indexed | 2024-03-10T01:45:59Z |
| publishDate | 2022-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Cancers |
| spelling | doaj.art-283959d772e2423b8b9b4163cdd957cc2023-11-23T13:15:10ZengMDPI AGCancers2072-66942022-01-0114244510.3390/cancers14020445A Few-Shot Learning Approach Assists in the Prognosis Prediction of Magnetic Resonance-Guided Focused Ultrasound for the Local Control of Bone Metastatic LesionsFang-Chi Hsu0Hsin-Lun Lee1Yin-Ju Chen2Yao-An Shen3Yi-Chieh Tsai4Meng-Huang Wu5Chia-Chun Kuo6Long-Sheng Lu7Shauh-Der Yeh8Wen-Sheng Huang9Chia-Ning Shen10Jeng-Fong Chiou11The Ph.D. Program for Translational Medicine, College of Medical Science and Technology, Taipei Medical University and Academia Sinica, Taipei 110, TaiwanDepartment of Radiology, School of Medicine, College of Medicine, Taipei Medical University, Taipei 110, TaiwanGraduate Institute of Biomedical Materials and Tissue Engineering, College of Biomedical Engineering, Taipei Medical University, Taipei 110, TaiwanDepartment of Pathology, School of Medicine, College of Medicine, Taipei Medical University, Taipei 110, TaiwanDepartment of Radiation Oncology, Shuang Ho Hospital, Taipei Medical University, Taipei 110, TaiwanDepartment of Orthopedics, Taipei Medical University Hospital, Taipei 110, TaiwanDepartment of Radiation Oncology, Taipei Medical University Hospital, Taipei 110, TaiwanDepartment of Radiation Oncology, Taipei Medical University Hospital, Taipei 110, TaiwanDepartment of Urology, School of Medicine, College of Medicine, Taipei Medical University, Taipei 110, TaiwanDepartment of Nuclear Medicine, Taipei Medical University Hospital, Taipei 110, TaiwanThe Ph.D. Program for Translational Medicine, College of Medical Science and Technology, Taipei Medical University and Academia Sinica, Taipei 110, TaiwanThe Ph.D. Program for Translational Medicine, College of Medical Science and Technology, Taipei Medical University and Academia Sinica, Taipei 110, TaiwanMagnetic resonance-guided focused ultrasound surgery (MRgFUS) constitutes a noninvasive treatment strategy to ablate deep-seated bone metastases. However, limited evidence suggests that, although cytokines are influenced by thermal necrosis, there is still no cytokine threshold for clinical responses. A prediction model to approximate the postablation immune status on the basis of circulating cytokine activation is thus needed. IL-6 and IP-10, which are proinflammatory cytokines, decreased significantly during the acute phase. Wound-healing cytokines such as VEGF and PDGF increased after ablation, but the increase was not statistically significant. In this phase, IL-6, IL-13, IP-10, and eotaxin expression levels diminished the ongoing inflammatory progression in the treated sites. These cytokine changes also correlated with the response rate of primary tumor control after acute periods. The few-shot learning algorithm was applied to test the correlation between cytokine levels and local control (<i>p</i> = 0.036). The best-fitted model included IL-6, IL-13, IP-10, and eotaxin as cytokine parameters from the few-shot selection, and had an accuracy of 85.2%, sensitivity of 88.6%, and AUC of 0.95. The acceptable usage of this model may help predict the acute-phase prognosis of a patient with painful bone metastasis who underwent local MRgFUS. The application of machine learning in bone metastasis is equivalent or better than the current logistic regression.https://www.mdpi.com/2072-6694/14/2/445magnetic resonance-guided focused ultrasound surgerymachine learningprognosis predictionbone metastasisHIFU |
| spellingShingle | Fang-Chi Hsu Hsin-Lun Lee Yin-Ju Chen Yao-An Shen Yi-Chieh Tsai Meng-Huang Wu Chia-Chun Kuo Long-Sheng Lu Shauh-Der Yeh Wen-Sheng Huang Chia-Ning Shen Jeng-Fong Chiou A Few-Shot Learning Approach Assists in the Prognosis Prediction of Magnetic Resonance-Guided Focused Ultrasound for the Local Control of Bone Metastatic Lesions Cancers magnetic resonance-guided focused ultrasound surgery machine learning prognosis prediction bone metastasis HIFU |
| title | A Few-Shot Learning Approach Assists in the Prognosis Prediction of Magnetic Resonance-Guided Focused Ultrasound for the Local Control of Bone Metastatic Lesions |
| title_full | A Few-Shot Learning Approach Assists in the Prognosis Prediction of Magnetic Resonance-Guided Focused Ultrasound for the Local Control of Bone Metastatic Lesions |
| title_fullStr | A Few-Shot Learning Approach Assists in the Prognosis Prediction of Magnetic Resonance-Guided Focused Ultrasound for the Local Control of Bone Metastatic Lesions |
| title_full_unstemmed | A Few-Shot Learning Approach Assists in the Prognosis Prediction of Magnetic Resonance-Guided Focused Ultrasound for the Local Control of Bone Metastatic Lesions |
| title_short | A Few-Shot Learning Approach Assists in the Prognosis Prediction of Magnetic Resonance-Guided Focused Ultrasound for the Local Control of Bone Metastatic Lesions |
| title_sort | few shot learning approach assists in the prognosis prediction of magnetic resonance guided focused ultrasound for the local control of bone metastatic lesions |
| topic | magnetic resonance-guided focused ultrasound surgery machine learning prognosis prediction bone metastasis HIFU |
| url | https://www.mdpi.com/2072-6694/14/2/445 |
| work_keys_str_mv | AT fangchihsu afewshotlearningapproachassistsintheprognosispredictionofmagneticresonanceguidedfocusedultrasoundforthelocalcontrolofbonemetastaticlesions AT hsinlunlee afewshotlearningapproachassistsintheprognosispredictionofmagneticresonanceguidedfocusedultrasoundforthelocalcontrolofbonemetastaticlesions AT yinjuchen afewshotlearningapproachassistsintheprognosispredictionofmagneticresonanceguidedfocusedultrasoundforthelocalcontrolofbonemetastaticlesions AT yaoanshen afewshotlearningapproachassistsintheprognosispredictionofmagneticresonanceguidedfocusedultrasoundforthelocalcontrolofbonemetastaticlesions AT yichiehtsai afewshotlearningapproachassistsintheprognosispredictionofmagneticresonanceguidedfocusedultrasoundforthelocalcontrolofbonemetastaticlesions AT menghuangwu afewshotlearningapproachassistsintheprognosispredictionofmagneticresonanceguidedfocusedultrasoundforthelocalcontrolofbonemetastaticlesions AT chiachunkuo afewshotlearningapproachassistsintheprognosispredictionofmagneticresonanceguidedfocusedultrasoundforthelocalcontrolofbonemetastaticlesions AT longshenglu afewshotlearningapproachassistsintheprognosispredictionofmagneticresonanceguidedfocusedultrasoundforthelocalcontrolofbonemetastaticlesions AT shauhderyeh afewshotlearningapproachassistsintheprognosispredictionofmagneticresonanceguidedfocusedultrasoundforthelocalcontrolofbonemetastaticlesions AT wenshenghuang afewshotlearningapproachassistsintheprognosispredictionofmagneticresonanceguidedfocusedultrasoundforthelocalcontrolofbonemetastaticlesions AT chianingshen afewshotlearningapproachassistsintheprognosispredictionofmagneticresonanceguidedfocusedultrasoundforthelocalcontrolofbonemetastaticlesions AT jengfongchiou afewshotlearningapproachassistsintheprognosispredictionofmagneticresonanceguidedfocusedultrasoundforthelocalcontrolofbonemetastaticlesions AT fangchihsu fewshotlearningapproachassistsintheprognosispredictionofmagneticresonanceguidedfocusedultrasoundforthelocalcontrolofbonemetastaticlesions AT hsinlunlee fewshotlearningapproachassistsintheprognosispredictionofmagneticresonanceguidedfocusedultrasoundforthelocalcontrolofbonemetastaticlesions AT yinjuchen fewshotlearningapproachassistsintheprognosispredictionofmagneticresonanceguidedfocusedultrasoundforthelocalcontrolofbonemetastaticlesions AT yaoanshen fewshotlearningapproachassistsintheprognosispredictionofmagneticresonanceguidedfocusedultrasoundforthelocalcontrolofbonemetastaticlesions AT yichiehtsai fewshotlearningapproachassistsintheprognosispredictionofmagneticresonanceguidedfocusedultrasoundforthelocalcontrolofbonemetastaticlesions AT menghuangwu fewshotlearningapproachassistsintheprognosispredictionofmagneticresonanceguidedfocusedultrasoundforthelocalcontrolofbonemetastaticlesions AT chiachunkuo fewshotlearningapproachassistsintheprognosispredictionofmagneticresonanceguidedfocusedultrasoundforthelocalcontrolofbonemetastaticlesions AT longshenglu fewshotlearningapproachassistsintheprognosispredictionofmagneticresonanceguidedfocusedultrasoundforthelocalcontrolofbonemetastaticlesions AT shauhderyeh fewshotlearningapproachassistsintheprognosispredictionofmagneticresonanceguidedfocusedultrasoundforthelocalcontrolofbonemetastaticlesions AT wenshenghuang fewshotlearningapproachassistsintheprognosispredictionofmagneticresonanceguidedfocusedultrasoundforthelocalcontrolofbonemetastaticlesions AT chianingshen fewshotlearningapproachassistsintheprognosispredictionofmagneticresonanceguidedfocusedultrasoundforthelocalcontrolofbonemetastaticlesions AT jengfongchiou fewshotlearningapproachassistsintheprognosispredictionofmagneticresonanceguidedfocusedultrasoundforthelocalcontrolofbonemetastaticlesions |