•Quantitative image analysis (QIA) is a suitable candidate for precision medicine.•Imaging quantitative features are highly sensitive to variations in image acquisition and reconstruction.•A ...framework for robust radiomics analysis is proposed.•Radiomics-specific harmonization methods are needed to meaningfully analyze the available retrospective imaging data.•A QIA-specific imaging is needed to ensure the generalizability of models developed.
The advancement of artificial intelligence concurrent with the development of medical imaging techniques provided a unique opportunity to turn medical imaging from mostly qualitative, to further quantitative and mineable data that can be explored for the development of clinical decision support systems (cDSS). Radiomics, a method for the high throughput extraction of hand-crafted features from medical images, and deep learning -the data driven modeling techniques based on the principles of simplified brain neuron interactions, are the most researched quantitative imaging techniques. Many studies reported on the potential of such techniques in the context of cDSS. Such techniques could be highly appealing due to the reuse of existing data, automation of clinical workflows, minimal invasiveness, three-dimensional volumetric characterization, and the promise of high accuracy and reproducibility of results and cost-effectiveness. Nevertheless, there are several challenges that quantitative imaging techniques face, and need to be addressed before the translation to clinical use. These challenges include, but are not limited to, the explainability of the models, the reproducibility of the quantitative imaging features, and their sensitivity to variations in image acquisition and reconstruction parameters. In this narrative review, we report on the status of quantitative medical image analysis using radiomics and deep learning, the challenges the field is facing, propose a framework for robust radiomics analysis, and discuss future prospects.
Several mutations in nuclear genes encoding for mitochondrial components have been associated with an increased cancer risk or are even causative, e.g. succinate dehydrogenase (SDHB, SDHC and SDHD ...genes) and iso-citrate dehydrogenase (IDH1 and IDH2 genes). Recently, studies have suggested an eminent role for mitochondrial DNA (mtDNA) mutations in the development of a wide variety of cancers. Various studies associated mtDNA abnormalities, including mutations, deletions, inversions and copy number alterations, with mitochondrial dysfunction. This might, explain the hampered cellular bioenergetics in many cancer cell types. Germline (e.g. m.10398A>G; m.6253T>C) and somatic mtDNA mutations as well as differences in mtDNA copy number seem to be associated with cancer risk. It seems that mtDNA can contribute as driver or as complementary gene mutation according to the multiple-hit model. This can enhance the mutagenic/clonogenic potential of the cell as observed for m.8993T>G or influences the metastatic potential in later stages of cancer progression. Alternatively, other mtDNA variations will be innocent passenger mutations in a tumor and therefore do not contribute to the tumorigenic or metastatic potential. In this review, we discuss how reported mtDNA variations interfere with cancer treatment and what implications this has on current successful pharmaceutical interventions. Mutations in MT-ND4 and mtDNA depletion have been reported to be involved in cisplatin resistance. Pharmaceutical impairment of OXPHOS by metformin can increase the efficiency of radiotherapy. To study mitochondrial dysfunction in cancer, different cellular models (like ρ0 cells or cybrids), in vivo murine models (xenografts and specific mtDNA mouse models in combination with a spontaneous cancer mouse model) and small animal models (e.g. Danio rerio) could be potentially interesting to use. For future research, we foresee that unraveling mtDNA variations can contribute to personalized therapy for specific cancer types and improve the outcome of the disease.
A straightforward way to improve local tumour control is to increase the radiation dose further, which is hampered by normal tissue constraints. Therefore, the concept of “dose painting” has been ...introduced, i.e. delivering a higher dose towards the more resistant tumour areas and reducing the dose towards the more sensitive area. Using this individualized strategy, the burden to the normal tissues can be kept constant; Probably the most investigated methodology for visualizing intra tumour radio‐resistance are hypoxia tracers. However, these tracers can have limitations, e.g. slow clearance of the unbound tracer in non‐hypoxic areas, low test re‐test performance, and low tumour to reference activity ratios can be observ. Also, as hypoxia is only one of several causes for radio‐resistance, there is a lot of uncertainty about how specific these hypoxia tracers relate to radio‐resistance;
Another important micro‐environmental factor is tumour glucose metabolism, commonly assessed by the glucose analogue 18F‐deoxyglucose (FDG) for PET imaging. FDG is the workhorse in the field of oncology for diagnostic and prognostic purposes that has led to improved clinical decision making in a large quantity of cancer patients. Many investigations have shown that high FDG uptake is prognostic for worse survival for patients both treated with radiotherapy and surgery. During the past decades FDG has become widely available and a lot of progress was made in standardisation and reproducibility of FDG‐PET imaging, further improving the usability in the clinic.
The underlying biology for FDG uptake is complex as it measures the uptake of glucose in cells by ATP independent glucose transporters. This makes FDG uptake specific to the metabolic pathway although this pathway is known to influence numerous other processes to facilitate cancer. Consequently, in contrast to hypoxia tracers, FDG uptake in the tumour does not reflect a single biologic characteristic, but is correlated to, although certainly not specific for, radio‐resistance, proliferation, cell density, hypoxia, mitochondrial dysfunction, and lipogenesis. Furthermore, in a pre‐clinical investigation, it has been shown that an increase of radiation dose showed a higher local control for tumours with higher FDG uptake compared to tumours with lower FDG uptake, affirming the relation between FDG‐uptake and radio‐resistance. Another argument is that the locations of the patterns of relapse are associated with the pre‐treatment high FDG uptake areas. Last year, our group has shown that the residual metabolic active areas after treatment show a high overlap with the high FDG uptake areas before treatment. This was confirmed in a study by the group of Dresden. These studies show that the location within the tumour with a high chance of metabolic relapse, and probably more radio‐resistant, can be identified before treatment using a single FDG‐PET‐CT scan. Furthermore, it was shown that these high FDG uptake areas remain stable during a course of radiotherapy, making it possible to keep a similar dose painting plan during treatment.
As FDG corresponds to several biologic characteristics important for radio‐ resistance, due to the confirmed relation with the locations of relapse, the wide availability, the standardized imaging protocols, and the stability of the uptake patterns during RT, FDG is an attractive tracer for dose painting. Due to these favourable characteristics and because of sufficient improvements in the accuracy of imaging and dose delivery techniques, we believe that the time is right to more towards carefully designed clinical trials, by redistributing the dose towards the high pre‐treatment FDG uptake areas. A randomized phase 2 trial in lung cancer started in 2010 in The Netherlands.
Basal cell carcinoma (BCC) is one of the most common types of cancer. The growing incidence worldwide and the need for fast, reliable and less invasive diagnostic techniques make a strong case for ...the application of different artificial intelligence techniques for detecting and classifying BCC and its subtypes. We report on the current evidence regarding the application of handcrafted and deep radiomics models used for the detection and classification of BCC in dermoscopy, optical coherence tomography and reflectance confocal microscopy. We reviewed all the articles that were published in the last 10 years in PubMed, Web of Science and EMBASE, and we found 15 articles that met the inclusion criteria. We included articles that are original, written in English, focussing on automated BCC detection in our target modalities and published within the last 10 years in the field of dermatology. The outcomes from the selected publications are presented in three categories depending on the imaging modality and to allow for comparison. The majority of articles (n = 12) presented different AI solutions for the detection and/or classification of BCC in dermoscopy images. The rest of the publications presented AI solutions in OCT images (n = 2) and RCM (n = 1). In addition, we provide future directions for the application of these techniques for the detection of BCC. In conclusion, the reviewed publications demonstrate the potential benefit of AI in the detection of BCC in dermoscopy, OCT and RCM.
Background: Patients with stage III non-small-cell lung cancer (NSCLC) and limited disease small-cell lung cancer are excluded from concurrent chemoradiation mostly on the basis of comorbidity and ...age. The purpose of this prospective study was to get insight in what proportion of patients with locally advanced lung cancer would be suitable for concurrent chemoradiation.
Patients and methods: From 2002 to 2005, all patients with a pathological diagnosis of lung cancer and with locally advanced disease in the Maastricht Cancer Registry, the Netherlands, comorbidity were prospectively assessed. Patients were regarded as noneligible for concurrent chemoradiation if they had one or more important comorbidity or were 75 years or older.
Results: In all, 711 patients were included, 577 with NSCLC and 134 with SCLC. Overall, 166 patients (23.3%) were 75 years or older. Of the 526 patients <75 years, comorbidities were as follows: 278 (52.9%) 0, 188 (35.7%) 1, and 56 (11.4%) 2 or more. In all, 408/686 (59%) of the whole patient group were considered as ineligible for concurrent chemoradiation.
Conclusions: More than half of patients with stage III lung cancer were theoretically not eligible for concurrent chemoradiation. Less toxic alternatives are needed for these patients.