Advanced survival prediction in head and neck cancer using hybrid machine learning systems and radiomics features

被引:2
|
作者
Salmanpour, Mohammadreza R. [1 ,2 ,3 ]
Hosseinzadeh, Mahdi [3 ,4 ]
Modiri, Ensiyeh [3 ,5 ]
Akbari, Azizeh [3 ,6 ]
Hajianfar, Ghasem [3 ,7 ]
Askari, Dariush [8 ]
Fatan, Mehdi [3 ]
Maghsudi, Mehdi [7 ]
Ghaffari, Hanieh [3 ]
Rezaei, Masoad [9 ]
Ghaemi, Mohammad M. [3 ,10 ]
Rahmim, Arman [1 ,2 ]
机构
[1] Univ British Columbia, Vancouver, BC, Canada
[2] BC Canc Res Inst, Vancouver, BC, Canada
[3] Technol Virtual Collaborat TECVICO Corp, Vancouver, BC, Canada
[4] Tarbiat Modares Univ, Tehran, Iran
[5] Islamic Azad Univ, Tehran, Iran
[6] Hakim Sabzevari Univ, Sabzevar, Iran
[7] Iran Univ Med Sci, Rajaie Cardiovasc Med & Res Ctr, Tehran, Iran
[8] Shahid Beheshti Univ Med, Tehran, Iran
[9] Ahvaz Jundishapur Univ Med Sci, Sch Med, Dept Med Phys, Ahvaz, Iran
[10] Kerman Univ Med Sci, Tehran, Iran
基金
加拿大自然科学与工程研究理事会;
关键词
survival prediction; hybrid machine learning system; radiomics features; fusion techniques; head and neck squamous cell carcinoma; image processing; dimension reduction algorithm; multicenter study;
D O I
10.1117/12.2612816
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
Accurate prognostic stratification of Head-and-Neck-Squamous-Cell-Carcinoma (HNSCC) patients can be an important clinical reference when designing therapeutic strategies. We set to predict 4 outcomes: overall survival (OS), distant metastasis (DM), locoregional recurrence (LR), and progression-free survival (SP). We studied Hybrid Machine Learning Systems (HMLS), applied to datasets with radiomics features. In this multicenter study, 408 HNSCC patients were extracted from The Cancer Imaging Archive (TCIA) database. PET images were registered to CT, enhanced, and cropped. 215 radiomics features were extracted from each region of interest via our standardized SERA radiomics package. We employed multiple HMLSs: 12 feature extraction ( FEA) or 9 feature selection algorithms (FSA) linked with 9 survival-prediction-algorithms (SPA) optimized by 5-fold cross-validation, applied to PET only, CT only and 4 PET-CT datasets generated by image-level fusion strategies. Datasets were normalized by z-score-technique, and c-indices were reported to compare the models. For OS prediction, the highest c-index 0.73 +/- 0.10 was obtained for HMLS with Ratio of low-pass pyramid (RP) fusion technique + gaussian process latent variable model ( GPLVM) + causal structure learning-based feature modification method (CSFM). For DM prediction, we achieved 0.80 +/- 0.06 via Dual-tree complex wavelet transform (DTCWT) fusion + Laplacian Score (LAP) + Logistic regression hazards (LH). For LR prediction, we arrived at a c-index of 0.73 +/- 0.13 using PET + Sammon Mapping Algorithm (SM)+ deep neural network to distribute first hitting times (DHS). For SP prediction, the performance of 0.68 +/- 0.02 was obtained via PET + SM + Relative risk model-depend on time (CoxTime). When no dimensionality reduction (FEA/FSA) was employed, the above 4 performances decreased to 0.69 +/- 0. 10, 0.74 +/- 0.13, 0.66 +/- 0.15, and 0.68 +/- 0.04 for OS, DM, LR and SP prediction. We demonstrated that using fusion techniques followed by appropriate HMLSs, including FEAs/FSAs and SPAs, improved prediction performance.
引用
收藏
页数:8
相关论文
共 50 条
  • [1] Prediction of TNM stage in head and neck cancer using hybrid machine learning systems and radiomics features
    Salmanpour, Mohammadreza R.
    Hosseinzadeh, Mahdi
    Akbari, Azizeh
    Borazjani, Kasra
    Mojallal, Kasra
    Askari, Dariush
    Hajianfar, Ghasem
    Rezaeijo, Seyed Masoud
    Ghaemi, Mohammad M.
    Nabizadeh, Amir Hossein
    Rahmim, Arman
    [J]. MEDICAL IMAGING 2022: COMPUTER-AIDED DIAGNOSIS, 2022, 12033
  • [2] Deep versus Handcrafted Tensor Radiomics Features: Prediction of Survival in Head and Neck Cancer Using Machine Learning and Fusion Techniques
    Salmanpour, Mohammad R.
    Rezaeijo, Seyed Masoud
    Hosseinzadeh, Mahdi
    Rahmim, Arman
    [J]. DIAGNOSTICS, 2023, 13 (10)
  • [3] Head and neck cancer treatment outcome prediction: a comparison between machine learning with conventional radiomics features and deep learning radiomics
    Huynh, Bao Ngoc
    Groendahl, Aurora Rosvoll
    Tomic, Oliver
    Liland, Kristian Hovde
    Knudtsen, Ingerid Skjei
    Hoebers, Frank
    van Elmpt, Wouter
    Malinen, Eirik
    Dale, Einar
    Futsaether, Cecilia Marie
    [J]. FRONTIERS IN MEDICINE, 2023, 10
  • [4] Deep learning of radiomics features for survival prediction in NSCLC and Head and Neck carcinoma
    Jochems, A.
    Hoebers, F.
    De Ruysscher, D.
    Leijenaar, R.
    Walsh, S.
    O'Sullivan, B.
    Bussink, J.
    Monshouwer, R.
    Leemans, R.
    Lambin, P.
    [J]. RADIOTHERAPY AND ONCOLOGY, 2017, 123 : S866 - S866
  • [5] Reliable and Reproducible Tensor Radiomics Features in Prediction of Survival in Head and Neck Cancer
    Salmanpour, M.
    Hosseinzadeh, M.
    Rezaeijo, S.
    Maghsudi, M.
    Rahmim, A.
    [J]. EUROPEAN JOURNAL OF NUCLEAR MEDICINE AND MOLECULAR IMAGING, 2022, 49 (SUPPL 1) : S20 - S20
  • [6] Tensor Deep versus Radiomics Features: Lung Cancer Outcome Prediction using Hybrid Machine Learning Systems
    Salmanpour, Mohammad R.
    Hosseinzadeh, Mahdi
    Sanati, Nima
    Jouzdani, Ali Fathi
    Gorji, Arman
    Mahboubisarighieh, Ali
    Maghsudi, Mehdi
    Rezaeijo, Seyed Masoud
    Moore, Sara
    Bonnie, Leung
    Uribe, Carlos
    Ho, Cheryl
    Rahmim, Arman
    [J]. JOURNAL OF NUCLEAR MEDICINE, 2023, 64
  • [7] Feasibility Of Prediction Of Radiation-Related Caries In Head-Neck Cancer Patients Using Machine Learning And Radiomics Features.
    Neves, L. V. F.
    Danelichen, A. F. B.
    Faustino, A.
    Matsuura, F. K.
    Viani, G.
    Azimbagirad, M.
    Faria, V. D. A.
    Pavoni, J. F.
    Felipe, J. C.
    Murta Junior, L. O.
    [J]. INTERNATIONAL JOURNAL OF RADIATION ONCOLOGY BIOLOGY PHYSICS, 2020, 108 (03): : E782 - E782
  • [8] Fusion-based tensor radiomics using reproducible features: Application to survival prediction in head and neck cancer
    Salmanpour, Mohammad R.
    Hosseinzadeh, Mahdi
    Rezaeijo, Seyed Masoud
    Rahmim, Arman
    [J]. COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE, 2023, 240
  • [9] Deep versus handcrafted tensor radiomics features: application to survival prediction in head and neck cancer
    Salmanpour, M.
    Hosseinzadeh, M.
    Rezaeijo, S.
    Ramezani, M.
    Marandi, S.
    Einy, M.
    Rahmim, A.
    [J]. EUROPEAN JOURNAL OF NUCLEAR MEDICINE AND MOLECULAR IMAGING, 2022, 49 (SUPPL 1) : S245 - S246
  • [10] Utility of adding Radiomics to clinical features in predicting the outcomes of radiotherapy for head and neck cancer using machine learning
    Gangil, Tarun
    Sharan, Krishna
    Rao, B. Dinesh
    Palanisamy, Krishnamoorthy
    Chakrabarti, Biswaroop
    Kadavigere, Rajagopal
    [J]. PLOS ONE, 2022, 17 (12):