As the most lethal major cancer, pancreatic cancer is a global healthcare challenge. Personalized medicine utilizing cutting-edge multi-omics data holds potential for major breakthroughs in tackling this critical problem. Radiomics and deep learning, two trendy quantitative imaging methods that take advantage of data science and modern medical imaging, have shown increasing promise in advancing the precision management of pancreatic cancer via diagnosing of precursor diseases, early detection, accurate diagnosis, and treatment personalization and optimization. Radiomics employs manually-crafted features, while deep learning applies computer-generated automatic features. These two methods aim to mine hidden information in medical images that is missed by conventional radiology and gain insights by systematically comparing the quantitative image information across different patients in order to characterize unique imaging phenotypes. Both methods have been studied and applied in various pancreatic cancer clinical applications. In this review, we begin with an introduction to the clinical problems and the technology. After providing technical overviews of the two methods, this review focuses on the current progress of clinical applications in precancerous lesion diagnosis, pancreatic cancer detection and diagnosis, prognosis prediction, treatment stratification, and radiogenomics. The limitations of current studies and methods are discussed, along with future directions. With better standardization and optimization of the workflow from image acquisition to analysis and with larger and especially prospective high-quality datasets, radiomics and deep learning methods could show real hope in the battle against pancreatic cancer through big data-based high-precision personalization.

(1) Background: Radiomics use high-throughput mining of medical imaging data to extract unique information and predict tumor behavior. Currently available clinical prediction models poorly predict treatment outcomes in pancreatic adenocarcinoma. Therefore, we used radiomic features of primary pancreatic tumors to develop outcome prediction models and compared them to traditional clinical models. (2) Methods: We extracted and analyzed radiomic data from pre-radiation contrast-enhanced CTs of 74 pancreatic cancer patients undergoing stereotactic body radiotherapy. A panel of over 800 radiomic features was screened to create overall survival and local-regional recurrence prediction models, which were compared to clinical prediction models and models combining radiomic and clinical information. (3) Results: A 6-feature radiomic signature was identified that achieved better overall survival prediction performance than the clinical model (mean concordance index: 0.66 vs. 0.54 on resampled cross-validation test sets), and the combined model improved the performance slightly further to 0.68. Similarly, a 7-feature radiomic signature better predicted recurrence than the clinical model (mean AUC of 0.78 vs. 0.66). (4) Conclusion: Overall survival and recurrence can be better predicted with models based on radiomic features than with those based on clinical features for pancreatic cancer.

To identify prognostic factors for grade 3 radiation dermatitis following passive-scattering proton therapy for breast cancer. This retrospective study included data on 23 (11 post-mastectomy and 12 post-lumpectomy) breast cancer patients who underwent proton therapy with the passive scattering technique in our institute from 2012 to 2016. Each patient received 50–50.4 cobalt Gy equivalent (CGE) at 1.8 or 2 CGE per daily fraction. Logistic regression analysis was performed to identify prognostic factors for grade 3 skin toxicity. Receiver operating characteristic (ROC) curve analysis and the area under the curve (AUC) were used to evaluate the performance of the models. 43% of the studied patients developed grade 3 radiation dermatitis. The dose-volume histogram (DVH) parameters of V52.5CGE and D10cm3 to skin5mm were correlated with grade 3 radiation dermatitis in both univariate and multivariate logistic regression analyses. Univariate logistic regression analysis suggested that D10cm3 to skin5mm (AUC = 0.69) and V52.5CGE to skin5mm (AUC = 0.70) were prognostic for grade 3 skin toxicity. The models using the combination of D10cm3 to skin5mm or V52.5CGE to skin5mm with breast volume marginally increased the AUC to 0.72 and 0.73, respectively. Models using the combination of D10cm3 to skin5mm or V52.5CGE to skin5mm with history of smoking increased the AUC to 0.75 and 0.83, respectively. In the current study, we identified prognostic factors for grade 3 radiation dermatitis in patients treated with passive-scattering proton therapy for breast cancer. This study provides promising tool for identifying high risk patients for whom treatment plan adjustment could be done to reduce the risk of radiation-induced grade 3 skin toxicity.

PB algorithms are commonly used for proton therapy. Previously reported limitations of the PB algorithm for proton therapy are mainly focused on high‐density gradients and small‐field dosimetry, the effect of PB algorithms on intensity‐modulated proton therapy (IMPT) for breast cancer has yet to be illuminated. In this study, we examined 20 patients with breast cancer and systematically investigated the dosimetric impact of MC and PB algorithms on IMPT. Four plans were generated for each patient: (a) a PB plan that optimized and computed the final dose using a PB algorithm; (b) a MC‐recomputed plan that recomputed the final dose of the PB plan using a MC algorithm; (c) a MC‐renormalized plan that renormalized the MC‐recomputed plan to restore the target coverage; and (d) a MC‐optimized plan that optimized and computed the final dose using a MC algorithm. The DVH on CTVs and on organ‐at‐risks (OARs) from each plan were studied. The Mann–Whitney U‐test was used for testing the differences between any two types of plans. We found that PB algorithms significantly overestimated the target dose in breast IMPT plans. The median value of the CTV D_99%, D_95%, and D_mean dropped by 3.7%, 3.4%, and 2.1%, respectively, of the prescription dose in the MC‐recomputed plans compared with the PB plans. The magnitude of the target dose overestimation by the PB algorithm was higher for the breast CTV than for the chest wall CTV. In the MC‐renormalized plans, the target dose coverage was comparable with the original PB plans, but renormalization led to a significant increase in target hot spots as well as skin dose. The MC‐optimized plans led to sufficient target dose coverage, acceptable target hot spots, and good sparing of skin and other OARs. Utilizing the MC algorithm for both plan optimization and final dose computation in breast IMPT treatment planning is therefore desirable.

Objective. To investigate synchrotron-based proton pencil beam scanning (PBS) beam delivery time (BDT) using novel continuous scanning mode. Approach. A BDT calculation model was developed for the Hitachi particle therapy system. The model was validated against the measured BDT of 36 representative clinical proton PBS plans with discrete spot scanning (DSS) in the current Hitachi proton therapy system. BDTs were calculated with the next generation using Mayo Clinic Florida system operating parameters for conventional DSS, and novel dose driven continuous scanning (DDCS). BDTs of DDCS with and without Break Spots were investigated. Main results. For DDCS without Break Spots, the use of Stop Ratio to control the transit dose largely reduced the beam intensity and consequently, severely prolonged the BDT. DDCS with Break Spots was able to maintain a sufficiently high beam intensity while controlling transit dose. In DDCS with Break Spots, tradeoffs were made between beam intensity and number of Break Spots. Therefore, BDT decreased with increased beam intensity but reached a plateau for beam intensity larger than 10 MU s ⁻¹ . Averaging over all clinical plans, BDT was reduced by 10% for DDCS with Break Spots compared to DSS. Significance. DDCS with Break Spots reduced BDT. DDCS has the potential to further reduce BDT under the ideal scenario which requests both stable beam intensity extraction and accurately modelling the transit dose. Further investigation is warranted.