Frequently Asked Questions

Q: Where is the name “Radiomics” coming from?

The suffix Omics is used to form nouns meaning a study of the totality of something; e.g. genomics, proteomics. In the case it refers to the totality of radiological images. Radiomics aims at the collective characterization and quantification of pools of image features.

Q: When was the name Radiomics used for the first time in a scientific publication?

In PubMed the first publications referring to “Radiomics” was the following:

Lambin P, Rios-Velazquez E, Leijenaar R, Carvalho S, van Stiphout RG, Granton P, Zegers CM, Gillies R, Boellard R, Dekker A, Aerts HJ Radiomics: extracting more information from medical images using advanced feature analysis. Eur J Cancer. 2012 Mar;48(4):441-6.

Q: What is the main result/finding being reported in the Nature Communications article and why is it so important?

  • Main result/finding:

    The CT scans which are used almost universally both in cancer drug development and in care of cancer patients contain a lot of information about the tumour (e.g. shape, texture) which has previously been discarded because people didn’t know how to quantitate it objectively. MAASTRO researchers have found out how to objectively quantify this information (i.e. craft it into a biomarker) and have shown that they can now predict outcome better than previously.
     
  • Why is it so important?

    Cancer patients who present similarly often have very different outcomes (e.g. survival). Stratified medicine aims to better predict individual patients’ outcomes, and better select optimal treatments, based on improved biomarkers.
     
  • Why is this finding important for patients?

    Stratified medicine in cancer patients is most powerful when based on molecular biomarkers extracted from the tumour. However for many cancer patients these data aren’t available – different parts of tumours have different molecular characteristics and anyway tumours change over time. It’s impossible to biopsy every part of every tumour every month so patients may be imperfectly characterised.
     
  • What is the importance of this finding for drug development?

    Stratified medicine has transformed cancer drug development. This completely independent approach will improve further that stratification leading potentially to smaller faster trials showing greater benefit which can justify higher reimbursement.

Q: The basic information (the underlying image) is fixed and will not change. Why/how do you get more information with Radiomics?

  •  We provide quantitative information (objective), that goes beyond simple measurements.
  •  It’s high throughput. Like a facial recognition software analyzing a crowd.
  •  Because Radiomics software can see information which is not seen and/or quantifiable by the human eye.

Q: Which types of images should be analyzed by Radiomics?

In our vision all the images produced by any hospital should be analyzed to produce more information and create or improve signatures.

Q: Does it work with any type of image?

In principle yes. At the moment we have strong data for both CT and PET images. With MR there is the issue of standardization between vendors. We can also use it for other types of images, like fluorescent, radiography, immunohistochemistry, pictures of skin lesions when taking into account calibration of these images.

Q: Is the medical community interested by Radiomics?

Yes, since our first paper (published in 2012), there is an exponential increase of papers (peer reviewed) referring to Radiomics (see PubMed). Several scientific meetings in the fields of radiology, radiotherapy, oncology now even have complete sessions on the topic.

Q: Are there published reviews on Radiomics?

Yes, the first paper mentioning Radiomics and explaining the hypothesis:
 
Lambin P, Rios-Velazquez E, Leijenaar R, Carvalho S, van Stiphout RG, Granton P, et al. Radiomics: extracting more information from medical images using advanced feature analysis. Eur J Cancer 2012;48: 441-446 (PubMed)
 
We are currently also writing an invited review for Nature Reviews Clinical Oncology.
 
Also see this excellent, recently published review paper by Gillies et al.:
 
Gillies RJ, Kinahan PE, Hricak H. Radiomics: Images Are More than Pictures, They Are Data. Radiology 2016;278: 563-577 (PubMed)
 
And the following paper by Kumar et al.:
 
Kumar V, Gu Y, Basu S, Berglund A, Eschrich SA, Schabath MB, et al. Radiomics: the process and the challenges. Magn Reson Imaging 2012;30: 1234-1248 (PubMed)

Q:  Is there a societal business case. Could the society save money with Radiomics?

  • Yes, in the field of diagnostics, by diagnosing cancer earlier and/or avoiding unnecessary biopsies by analyzing in a quantitative way suspicious lesions.
  • In the field of theragnostics, by stratifying patients with easy to perform imaging biomarkers, to the right treatment.
  • Overall radiomics is cheaper then genomics or proteomic

Q: What is the role of patients?

We envision Radiomics to be used in the context of shared clinical decision making. In a few years a radiomic “sequence” will be available for all patients.

Q: Will the radiologist be threatened?

  • No, we see Radiomics as a means to help the radiologist make more informed decisions by providing objective, quantitative and robust information.
  • By studying all the structures visible on an image we take away a large portion of the workload.

Q: Are there applications for Radiomics outside radiotherapy/radiology?

Yes, we see possible applications of Radiomics outside the fields of radiotherapy and radiology. An example can be given for dermatology, where premalignant naevi could be identified, in high-throughput, also looking at their change, with a Radiomics approach.

 

Q: Will you update your signatures, and how?

Yes, we will update our signatures with new data available for research. We will furthermore exploit DISTRIM, our distributed learning solution to update our signatures (click here to watch the animation).

 

Q: How would you rank your research?

We rank our research in the field of Radiomics in the top 10%, as is proven by high impact papers (see PubMed).

Is there enough scientific evidence in favor of the radiomics software to justify its use in the care?

Yes, we believe we have enough scientific evidence that this software is working. Usefulness of the software has been tested in six clinical data sets coming from different centers, comprising imaging data of 930 patients with lung or head and neck cancer. We found a large number of radiomic features, determined by the software, to have strong prognostic power. A multi-feature radiomic signature capturing intra-tumor heterogeneity, was identified and independently validated as being strongly prognostic for overall survival in both lung and head and neck cancer. We believe there are few new molecular biomarkers with such an extensive clinical validation. What makes this approach attractive is that it is non invasive and very cheap because we use existing standard imaging.

Q: What could be the clinical applications of the radiomics software?

The main advantage, at the moment, would be for inoperable non-metastatic non-small cell lung cancer, which is known to be a very prominent disease. We know that the TNM classification, which is the base of all the existing guidelines, is not predictive at all to make the crucial decision between palliative and curative treatment. For example, there are T4 and N3 patients with good prognosis that could be selected for curative treatment even though the guidelines propose a palliative treatment. We also see a clear indication for the growing elderly population, which are usually undertreated, for which we could help to select patients with good prognosis without extra invasive maneuvers. In short, this non-invasive radiomics tool could help the multidisciplinary team to make decisions in the treatment of inoperable non-small cell lung cancer and more specifically help to avoid under treatment.

Q: Is it possible to obtain a license for the radiomics software? And for signature(s) as well?

Yes, licensing is possible. If you are interested in licensing options, please use the contact form or contact information at ptTheragnostic.