Artificial intelligence in cancer imaging: Clinical challenges and applications

Abstract

Judgement, as one of the core tenets of medicine, relies upon the integration of multilayered data with nuanced decision making. Cancer offers a unique context for medical decisions given not only its variegated forms with evolution of disease but also the need to take into account the individual condition of patients, their ability to receive treatment, and their responses to treatment. Challenges remain in the accurate detection, characterization, and monitoring of cancers despite improved technologies. Radiographic assessment of disease most commonly relies upon visual evaluations, the interpretations of which may be augmented by advanced computational analyses. In particular, artificial intelligence (AI) promises to make great strides in the qualitative interpretation of cancer imaging by expert clinicians, including volumetric delineation of tumors over time, extrapolation of the tumor genotype and biological course from its radiographic phenotype, prediction of clinical outcome, and assessment of the impact of disease and treatment on adjacent organs. AI may automate processes in the initial interpretation of images and shift the clinical workflow of radiographic detection, management decisions on whether or not to administer an intervention, and subsequent observation to a yet to be envisioned paradigm. Here, the authors review the current state of AI as applied to medical imaging of cancer and describe advances in 4 tumor types (lung, brain, breast, and prostate) to illustrate how common clinical problems are being addressed. Although most studies evaluating AI applications in oncology to date have not been vigorously validated for reproducibility and generalizability, the results do highlight increasingly concerted efforts in pushing AI technology to clinical use and to impact future directions in cancer care.

Repeatability of radiomics and machine learning for DWI: Short‐term repeatability study of 112 patients with prostate cancer

“…Some of the conventional radiomics, such as texture methods, demonstrated high AUC but low repeatability, stressing the fact that high classification potential in training set does not necessarily mean good overall performance, specifically for classifying PCa between GGG 1 versus >1. Therefore, features with low repeatability but high classification performance for PCa classification should be considered with caution, as the feature may turn out to have poor short‐term repeatability, whereas high short‐term repeatability is needed for practical application of them in non‐invasive PCa detection, characterization, therapy planning, and therapy monitory …”

“…Qualitative evaluation by means of visual inspection and interpretation of the medical images is a routine clinical practice, whereas image‐derived measurements have been shown to be promising aides to a radiologist in lesion detection and characterization . During the last decade, radiomics, including textures and machine learning (ML), have been applied extensively in medical imaging in general . Adoption of these methods in routine clinical practice has been limited by concerns related to poor repeatability and reproducibility because of different degree of noise in the imaging data sets obtained from different imaging sessions .…”

“…1 During the last decade, radiomics, [2][3][4] including textures and machine learning (ML), have been applied extensively in medical imaging in general. [5][6][7] Adoption of these methods in routine clinical practice has been limited by concerns related to poor repeatability and reproducibility 8 because of different degree of noise in the imaging data sets obtained from different imaging sessions. 9 An increasing number of research groups are applying radiomics for prostate cancer (PCa) imaging.…”

Additive Manufacturing of Precision Biomaterials

“…The ability to harness these data sets for the design of precision biomaterials will be critical to its successful implementation. Machine learning is already implemented in the assessment and diagnosis of high content medical images . Adapting these skills to anatomical images during surgical planning or device design will enable machine learning to assist in the optimization of the structure, materials, and print path for the AM of biomedical devices.…”

CT radiomics can help screen the Coronavirus disease 2019 (COVID-19): a preliminary study

“…Radiomics, a quantitative analysis technology based on medical imaging, has been widely used in the oncology research [17][18][19]. Previous radiomics studies showed that the quantitative radiomic features could represent the changes in pathology and gene level, and thus had encouraging performance in cancer diagnosis, treatment outcome prediction, and prognosis prediction [20][21][22][23]. Besides application in cancer, radiomics has also been used in the heart diseases, such as hypertrophic cardiomyopathy [24] and coronary heart disease [25].…”