|Year : 2022 | Volume
| Issue : 1 | Page : 33-35
Image “OMICS” in cancer - Synergy of qualitative and quantitative data analytics for futuristic precision medicine
G Lohith1, Kritika Murugan2, Krithikaa Sekar1, Sudhakar Sampangi3, Mahesh Bandemagal2, Shivakumar Swamy3
1 Department of Radiation Oncology, HealthCare Global (HCG) Cancer Centre, Bengaluru, Karnataka, India
2 Department of Surgical Oncology, HealthCare Global (HCG) Cancer Centre, Bengaluru, Karnataka, India
3 Department of Radiology, HealthCare Global (HCG) Cancer Centre, Bengaluru, Karnataka, India
|Date of Submission||18-Sep-2021|
|Date of Decision||08-Mar-2022|
|Date of Acceptance||11-Mar-2022|
|Date of Web Publication||03-May-2022|
Dr. Kritika Murugan
Consultant Surgical Oncologist, Department of Surgical Oncology, HealthCare Global (HCG) Cancer Centre, Bengaluru, Karnataka
Source of Support: None, Conflict of Interest: None
There are several methods for generating and analyzing big data in oncology, the most well-known of which are genomics, proteomics, and metabolomics. Similarly, “omics” clusters in imaging are frequently referred to as “radiomics.” A quantitative approach to medical imaging that tries to improve current qualitative data with modern computation and often counterintuitive mathematical analysis. This paper describes the breakthroughs in the use of radiomics in breast cancer, as well as the future challenges of radiomics research.
Keywords: Medical imaging, omics, oncology, precision diagnosis and treatment, radiomics
|How to cite this article:|
Lohith G, Murugan K, Sekar K, Sampangi S, Bandemagal M, Swamy S. Image “OMICS” in cancer - Synergy of qualitative and quantitative data analytics for futuristic precision medicine. J Precis Oncol 2022;2:33-5
|How to cite this URL:|
Lohith G, Murugan K, Sekar K, Sampangi S, Bandemagal M, Swamy S. Image “OMICS” in cancer - Synergy of qualitative and quantitative data analytics for futuristic precision medicine. J Precis Oncol [serial online] 2022 [cited 2022 Nov 29];2:33-5. Available from: https://www.jprecisiononcology.com//text.asp?2022/2/1/33/344533
| Introduction|| |
In oncology, different ways to generate and evaluate big data exist, most commonly known fields of genomics, proteomics, or metabolomics. Similarly, “omics” clusters in imaging have been used increasingly called “radiomics.” Quantitative approach to medical imaging, which aims at enhancing the existing qualitative data by means of advanced computing and sometimes nonintuitive mathematical analysis. Radiomics, most extensively has been applied in the domain of oncology, is based on the assumption that biomedical images contain information of disease-specific processes through quantitative analytics. Extraction of the spatial distribution of signal intensities and pixel interrelationships, radiomics quantifies textural information by mathematical and statistical modeling using advanced computation and artificial intelligence software. The subjective nature of image interpretation such as image intensity, shape, or texture can be quantified by means of radiomics, thus does not imply any automation of the diagnostic processes, rather it provides augmentation of existing ones with additional data., Radiomics is well placed to make clinical and cost-effective contributions to cancer care as a decision-making tool for precision medicine. However, a systematic reproducible framework needs to be established so that the benefits can be demonstrated with confidence. This article reviews the advances in the application of radiomics in breast cancer and the future challenges of research in radiomics.
| Integration of Radiomics to Precision Oncology for Advancement of Care|| |
Radiomics in combination with other “-omics” studies, with the advancement of bioinformatics, systems applications with high artificial intelligence capabilities have made precision oncology a reality and helped in progressive research to combine imaging to map the tumor possibilities for appropriate treatment, prediction, and prognostication.
Imaging with rich spatial and temporal information has advantages over other genomics, proteomics, or massive data, which is more effective in mapping the complete heterogeneity of the tumor with the usage of appropriate imaging technology which is noninvasive in nature and also in giving a comprehensive tumor factor in combination with other “omics” data for better outcomes in cancer care.
| Future Challenges of Radiomics|| |
Since the origin of radiomics in 2012, year-on-year, there as much exciting research done the field with positive outcome but also more and more limitations have been also been exposed, including reproducibility of the results and also the image acquisition protocols are different in different institutions and hence the usage of radiomic results have been difficult in the clinical scenario, a brief highlight on the mitigation of challenges are presented below for future adoption in the clinics.
| Images Acquisition and Reconstruction|| |
Imaging equipment, quantitative normalization parameters evaluation, and reconstruction methodologies are varied in medical centers and due to this variability the stability of the radiomic features may be affected and this is because of lack of uniform protocol across centers for image acquisition and processing which can alter with subtle differences in image quality causing a difference in evaluating the radiomic quantitative features. The parameters for image acquisition must be uniform and be strictly controlled; otherwise, the normalization and other procedures should be applied to image preprocessing.,,
| Images Segmentation|| |
At present, manual segmentation is considered to be the standard, however, it is associated with subjectivity and is time-consuming. The development of semi-automated or automated segmentation wizards that are uniform and have less observer bias is an important step in the image segmentation area. This allows for reproducibility, which makes features of the software robust, accurate, and efficient across centers, decreasing uncertainty.,
| Features Extraction|| |
A large number of features extracted in radiomics results the need to reduce the data dimensions, which in turn helps in reproducibility and also in nonrepetition of extracted features. The flawed idea of having more features help in developing radiomic algorithms more robust should be ignored and instead less features with the least redundancy should be incorporated for better feature selection to develop and answer the intended endpoint. Second, various extraction methods, on how to standardize the data in the process of reducing redundancy are also a challenge.
There needs to be a library to develop a valuable extraction criterion for radiomics, “semantic” and “agnostic” features are the two types of features extracted in radiomics. Semantic features are those commonly used in the radiology lexicon to describe regions of interest, while agnostic features are those that attempt to capture lesion heterogeneity through quantitative descriptors. Radiomics needs large datasets for more robust classifiers and models but can be done with 100 datasets as well.
| Integration of Other “Omics” or Clinical Data|| |
Demographic, clinical, and other “Omics” data integrated with stable, robust machine learning, deep learning algorithms, or statistical methods to establish classification models or a prediction model to be conducive and transformed to clinical applications. However, as described previously having more features with a limited sample size maybe suboptimal for a robust model to be used in clinical valuation and most research presently existent in radiomics are single-institution studies and hence the availability of open-source imaging database with clinical and genomic data could be used to develop conclusions and also to use the data for external validation of the model developed. Radiomics studies must be repeatedly tested, refined made reproducible, and externally validated by multicenter, large sample, and randomized controlled clinical trials. This will enable us to guide clinical treatment accurately, reliably, and effectively.
| Conclusion|| |
Artificial intelligence and big data can be used to greatly promote the development of individualized treatment strategies. Every process of radiomics is faced with challenges, but the challenges often coexist with the opportunities. The future is the era of artificial intelligence-precision medicine, and we have many clinical resources to utilize. There is still a long way to go to truly achieve a personalized, precise treatment of cancer.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Neisius U, El-Rewaidy H, Nakamori S, Rodriguez J, Manning WJ, Nezafat R. Radiomic analysis of myocardial native T1
imaging discriminates between hypertensive heart disease and hypertrophic cardiomyopathy. JACC Cardiovasc Imaging 2019;12:1946-54.
Castellano G, Bonilha L, Li LM, Cendes F. Texture analysis of medical images. Clin Radiol 2004;59:1061-9.
Tourassi GD. Journey toward computer-aided diagnosis: Role of image texture analysis. Radiology 1999;213:317-20.
Mazurowski MA. Radiogenomics: What it is and why it is important. J Am Coll Radiol 2015;12:862-6.
Mackin D, Fave X, Zhang L, Fried D, Yang J, Taylor B, et al.
Measuring computed tomography scanner variability of radiomics features. Invest Radiol 2015;50:757-65.
Larue RT, van Timmeren JE, de Jong EE, Feliciani G, Leijenaar RT, Schreurs WM, et al.
Influence of gray level discretization on radiomic feature stability for different CT scanners, tube currents and slice thicknesses: A comprehensive phantom study. Acta Oncol 2017;56:1544-53.
Shafiq-Ul-Hassan M, Zhang GG, Latifi K, Ullah G, Hunt DC, Balagurunathan Y, et al.
Intrinsic dependencies of CT radiomic features on voxel size and number of gray levels. Med Phys 2017;44:1050-62.
Qiu Q, Duan J, Duan Z, Meng X, Ma C, Zhu J, et al
. Reproducibility of radiomic features with GrowCut and GraphCut semiautomatic tumor segmentation in hepatocellular carcinoma. Transl Cancer Res 2017;6:940-8.
Qiu Q, Duan J, Duan Z, Meng X, Ma C, Zhu J, et al.
Reproducibility and non-redundancy of radiomic features extracted from arterial phase CT scans in hepatocellular carcinoma patients: Impact of tumor segmentation variability. Quant Imaging Med Surg 2019;9:453-64.
Haralick R, Shanmugan K, Dinstein I. Textural features for image classification. IEEE Trans Syst Man Cybern 1973;3:610-21.
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;48:441-6.
Hassanpour S, Langlotz CP. Information extraction from multi-institutional radiology reports. Artif Intell Med 2016;66:29-39.