Learning objectives
To appreciate conditions visible in chest CTs where a radiomics approach may discover predictive markers.
To know the current gaps in knowledge and determine the next steps to apply radiomics to routine data.
To become familiar with data sources that add to the practicability of radiomics studies.
Background
Radiomics is an emerging field that harvests comprehensive multi-variate information from images to predict disease course and / or prognosis,
and is predominantly used in oncology (1-3).
For example,
comorbidities, such as Chronic Obstructive Pulmonary Disease (COPD) or congestive heart disease,
have a major impact on survival in lung cancer (4,
5).
Therefore,
a diagnostic approach that includes only tumor-specific factors falls short of a comprehensive characterization of all the individual traits of a patient,
which is necessary for personalized-medicine (6).
With the emergence of...
Findings and procedure details
What is radiomics?
Radiomics is the high-throughput extraction and analysis of quantitative imaging features from medical images (1,
8,
9).
There are different groups of features (e.g.,
first-order statistics,
shape- and size-based features,
textural features,
and wavelet features) and each group utilizes a specific mathematical approach to provide information about the image.
Whereas textural features were already used decades ago to analyze chest radiographs (10),
the rapid advances in machine-learning techniques have opened up opportunities in image analysis that surpass previous achievements every year.
It...
Conclusion
The application of radiomics in diseases other than tumours is possible and opens new predictive potential.
Moreover,
radiomics may provide quantitative information to radiologists to support the finding of the correct diagnosis.
Due to the complexity of the workflow,
inter-professional collaboration and the efficient use of data sources are necessary.
Utilizing large screening cohorts and existing data repositories may facilitate the development of radiomics and enable its implementation for the evaluation of routine data.
Personal information
S.
Röhrich.Department of Biomedical Imaging and Image-guided Therapy,
Medical University of Vienna,
Vienna,
Austria.
H.
Prosch.
Department of Biomedical Imaging and Image-guided Therapy,
Medical University of Vienna,
Vienna,
Austria.
W.H.
Sommer.
Department of Clinical Radiology,
LMU University of Munich,
Munich,
Germany.
K.M.
Thierfelder.
Department of Clinical Radiology,
LMU University of Munich,
Munich,
Germany.
G.
Langs.
Department of Biomedical Imaging and Image-guided Therapy,
CIR Lab,
Medical University of Vienna,
Vienna,
Austria
References
1. Gillies RJ,
Kinahan PE,
Hricak H.
Radiomics: Images Are More than Pictures,
They Are Data.
Radiology.
2015:151169.
2. Yip SS,
Aerts HJ.
Applications and limitations of radiomics.
Physics in medicine and biology.
2016;61(13):R150-66.
3. Lee G,
Lee HY,
Park H,
Schiebler ML,
van Beek EJ,
Ohno Y,
et al.
Radiomics and its emerging role in lung cancer research,
imaging biomarkers and clinical management: State of the art.
European journal of radiology.
2016.
4. Islam KM,
Jiang X,
Anggondowati T,
Lin G,
Ganti AK.
Comorbidity...
