Radiomic-based deep learning for time-to-event outcome in pulmonary malignancies - ON-115

Preferred Disciplines and Level: Software engineer/ Machine Learner (Masters, PhD, Post-Doc) 
Company: 16 Bit
Project Length: 8-12 months (1-2 units)
Desired start date: ASAP
Location: Toronto, Ontario
No. of Positions: 1
Preferences: Language: English / French          

About the Company: 

16 Bit is a Toronto-based AI-first medical imaging company with the vision of improving the quality and accessibility of healthcare through artificially intelligent tools. 16 Bit recently earned 1st place at the 2017 RSNA machine learning competition for their cutting edge pediatric bone age algorithm, which is being deployed to hundreds of U.S hospitals. As radiologists and computer scientists, the founders of 16 Bit are best positioned to develop, iterate, and deploy clinically useful AI radiology solutions that will empower physicians and allow them to provide the best possible care to their patients. 

Project Description:

Stereotactic body radiation therapy (SBRT) delivers high dose of radiation with high precision limiting the effect to adjacent tissue to a minimum and is broadly used in the treatment of primary and secondary pulmonary malignanices as an alternative to surgical resection. However, the criteria to select the patients and the lesions to treat are clinical and morphological, and the likelihood of local or distant failure for each lesion cannot be reliably assessed previous to the therapy.

Radiomics is the conversion of digital images into a high-dimensional data, motivated by the concept that biomedical images contain information that reflects underlying pathophysiology and tumor heterogeneity and that these relationships can be revealed via quantitative image analyses. In radiomics, feature selection is used to identify prognostic biomarkers (signature). Machine learning algorithms can subsequently use the signature to construct predictive models by learning the decision boundaries of the underlying data distribution.

The aim of this project is to develop a radiomics-based machine learning pipeline based on pre-treatment medical images of pulmonary tumors treated with SBRT for prediction of time-to-event clinical outcome.

Research Objectives:​

  • Development of radiomics analysis signature based on pre-SBRT treated pulmonary malignancies
  • Apply machine learning methods to build radiomics-based predictive models for the prediction of clinical outcomes


To be determined

Expertise and Skills Needed:

  • Experience with Python
  • Experience with MatLab coding
  • Previous experience with radiomics would be an asset
  • Experience with time-to-event statistical analysis
  • Experience with machine learning classifiers
  • Understanding of professional software development and design practices


For more info or to apply to this applied research position, please

  1. Check your eligibility and find more information about open projects.

  2. Interested students need to get the approval from their supervisor and send their CV along with a link to their supervisor’s university webpage by applying through the webform or directly to Camila Londono at clondono(a)