Prediction of therapy outcome

IMPACT Team: Image and models for the prediction of therapy outcome


The increase in life expectation combined with screening and early diagnosis of pathologies, for some of them, make more than ever arise the need to predict the consequences of treatment options to improve the therapy outcome, reduce the risk of side effects and complications.

Several studies were conducted within a classical methodological framework to identify predictors of clinical outcomes for different therapies : cardiac and vascular surgery, interventional cardiology and rythmology, urology, radiation oncology.

Thanks to advances in computer technology, predictive models know a renewed interest and constitute a new research evolving field, especially in radiation therapy. Our work in this field covered the use of approaches based on simple dose descriptors (such as the dose-volume histogram) and patient clinical history, such as logistic regression, nomograms and classic NTCP models, but also the development of new prediction models based on those imput variables (such as random forest and independent principal components) and  the study of the dose distribution witihin a population framework.

Prediction of toxicity
Segmental maping of the rectum to identify a sub-region at risks of rectal bleeding

Radiation therapy and oncology, some recent publications:

 

 

 

  • Fardoun, T., Chaste, D., Oger, E., Mathieu, R., Peyronnet, B., Rioux-Leclercq, N., Verhoest, G., Patard, J.J. and Bensalah, K., Predictive factors of hemorrhagic complications after partial nephrectomy. Eur J Surg Oncol. 2014;40(1):85-9.
  • Zhu JA, Zhang ZC, Li BS, Liu M, Yin Y, Yu JM, Luo LM, Shu HZ, De Crevoisier R. Analysis of Acute Radiation-Induced Esophagitis in Non-Small-Cell Lung Cancer Patients Using the Lyman Ntcp Model. Radiotherapy and Oncology. 2010;97(3):449-54.
  •  Mathieu R, Ospina JD, Beckendorf V, Delobel JB, Messai T, Chira C, Bossi A, Le Prise E, Guerif S, Simon JM, Dubray B, Zhu J, Lagrange JL, Pommier P, Gnep K, Acosta O, De Crevoisier R. Nomograms to Predict Late Urinary Toxicity after Prostate Cancer Radiotherapy. World Journal of Urology. 2014;32(3):743-51.
  • Ospina JD, Zhu J, Chira C, Bossi A, Delobel JB, Beckendorf V, Dubray B, Lagrange JL, Correa JC, Simon A, Acosta O, de Crevoisier R. Random Forests to Predict Rectal Toxicity Following Prostate Cancer Radiation Therapy. Int J Radiat Oncol Biol Phys. 2014;89(5):1024-31.
  • Acosta O, Dowling J, Drean G, Simon A, Crevoisier R, Haigron P. Multi-Atlas-Based Segmentation of Pelvic Structures from CT Scans for Planning in Prostate Cancer Radiotherapy. Abdomen and Thoracic Imaging, El-Baz AS, Saba L, Suri J, eds. Springer US. 2014:623-56.
  • Acosta O, Drean G, Ospina JD, Simon A, Haigron P, Lafond C, de Crevoisier R. Voxel-Based Population Analysis for Correlating Local Dose and Rectal Toxicity in Prostate Cancer Radiotherapy. Physics in Medicine and Biology. 2013;58(8):2581-95.
  • Ospina Arango J, Commandeur F, Rios R, Dréan G, Correa JC, Simon A, Haigron P, de Crevoisier R, Acosta O. A Tensor-Based Population Value Decomposition to Explain Rectal Toxicity after Prostate Cancer Radiotherapy. Proceedings of the 16th International Conference Medical Image Computing and Computer-Assisted Intervention – MICCAI 2013; 2013 22-26 sept; Nagoya, Japan.  p. 387-94.
  •  Coloigner, J., A. Fargeas, A. Kachenoura, L. Wang, G. Drean, C. Lafond, L. Senhadji, R. de Crevoisier, O. Acosta and L. Albera. A Novel Classification Method for Prediction of Rectal Bleeding in Prostate Cancer Radiotherapy Based on a Semi-Nonnegative ICA of 3D Planned Dose Distributions. IEEE Journal of Biomedical and Health Informatics. 2015;19(3):1168-1177.
  • Fargeas A, Albera L, Kachenoura A, Drean G, Ospina JD, Coloigner J, Lafond C, Delobel JB de Crevoisier R, Acosta O. "On feature extraction and classification in prostate cancer radiotherapy using tensor decompositions." Medical Engineering and Physics, Volume 37, Issue 1, January 2015, Pages 126–131, DOI: 10.1016/j.medengphy.2014.08.009

 

Cardiovascular interventions and surgery, some recent publications:

 

  • - Kaladji, A., Cardon, A., Abouliatim, I., Campillo-Gimenez, B., Heautot, J.F. and Verhoye, J.-P., Preoperative predictive factors of aneurysmal regression using the reporting standards for endovascular aortic aneurysm repair. Journal of Vascular Surgery. 2012;55(5):1287-1295.
  • - Auffret, V., Boulmier, D., Oger, E., Bedossa, M., Donal, E., Laurent, M., Sost, G., Beneux, X., Harmouche, M., Verhoye, J.P. and Le Breton, H., Predictors of 6-month poor clinical outcomes after transcatheter aortic valve implantation. Arch Cardiovasc Dis. 2014;107(1):10-20.
  • - Mascle S, Schnell F, Thebault C, Corbineau H, Laurent M, Hamonic S, Veillard D, Mabo P, Leguerrier A, Donal E. Predictive value of global longitudinal strain in a surgical population of organic mitral regurgitation. J Am Soc Echocardiogr. 2012;25(7):766-72.
  • - Harmouche M, Maasrani M, Corbineau H, Verhoye JP, Drochon A. A More Sensitive Pressure-Based Index to Estimate Collateral Blood Supply in Case of Coronary Three-Vessel Disease. Medical Hypotheses. 2012;79(2):261-3.