Treatment data and technical process challenges for practical big data efforts in radiation oncology

Abstract

The term Big Data has come to encompass a number of concepts and uses within medicine. This paper lays out the relevance and application of large collections of data in the radiation oncology community. We describe the potential importance and uses in clinical practice. The important concepts are then described and how they have been or could be implemented are discussed. Impediments to progress in the collection and use of sufficient quantities of data are also described. Finally, recommendations for how the community can move forward to achieve the potential of big data in radiation oncology are provided.

Head and Neck Cancer Adaptive Radiation Therapy (ART): Conceptual Considerations for the Informed Clinician

“…As the field of radiotherapy has become increasing more complex in terms of the increasing number and complexity of disparate information sources which must be aggregated to extract meaningful data, ART represents, in some sense, the index case of an unmet need for "Big Data" information support. This is both for ensuring patient safety and for correlating image-dose-response data in a clinical utilizable manner (51), because the volume, temporal, and spatial correlation of multiple elements (patient data, accumulated dose, images, DVH, DVF, positional shifts, toxicity/outcome) must be carefully collated, organized, curated, recorded and reported (52)(53)(54)(55)(56)(57). However, if the present situation persists, it will remain, as it is currently, almost impossible to effectively reconstruct an institutions' specific adaptive protocol in the absence of identical vendor-supplied treatment planning, registration, archiving, electronic medical record, toxicity and patient-reported outcomes collection, and outcome monitoring, barring significant resource allocation (58).…”

Combination of a Big Data Analytics Resource System With an Artificial Intelligence Algorithm to Identify Clinically Actionable Radiation Dose Thresholds for Dysphagia in Head and Neck Patients

“…Austria, version 4.3.3). [2][3][4][5][6][7][8][9][10][11][12][13][14][15] For each SSMS, a receiver operator characteristic curve was constructed, and the area under the curve (AUC) was calculated for each set of toxicity and DVH metric dose records. A DVH metric value threshold was determined with the Youden index and used to construct a 2 Â 2 contingency table.…”

“…Recently, we have constructed a big data analytics resource system (BDARS) that automates aggregation, integration, and harmonization of key data elements and relationships for all treated patients in a standardized framework. 7,8 Aggregated elements include dose volume histograms (DVHs) for all treated plans and the course cumulative as treated plan sum in both physical (Gy) and bio-corrected (equivalent dose [EQD2] Gy with a/b Z 2.5, 5, 10) doses. 8,9 Common Terminology Criteria for Adverse Events toxicity grades were entered in our electronic health record (Epic, Verona, WI) using standardized smart list objects we developed to enable accurate, automated extraction from encounter notes with aggregation into our BDARS.…”

“…7,8 Aggregated elements include dose volume histograms (DVHs) for all treated plans and the course cumulative as treated plan sum in both physical (Gy) and bio-corrected (equivalent dose [EQD2] Gy with a/b Z 2.5, 5, 10) doses. 8,9 Common Terminology Criteria for Adverse Events toxicity grades were entered in our electronic health record (Epic, Verona, WI) using standardized smart list objects we developed to enable accurate, automated extraction from encounter notes with aggregation into our BDARS. 10 Our objective in this study was to develop an automatable, systematic approach that enabled consideration of both physical and biologically corrected doses to both percentage and absolute volumes of organs at risk, detailing levels of evidence for each candidate metric.…”

Radiation Therapy Quality Assurance Tasks and Tools: The Many Roles of Machine Learning

“…Even if we could construct a model that captures all related variables, it can be challenging to accurately populate models when some amounts of training data contain missing values. The fact that there are data elements that do not require entry to enable treatment (and are thus left out of the R&V system or OIS), in addition to database areas where data structure is nonstandardized, has lead to data missing during automated extraction of patient data from OIS and EHR . Although the absence of these kinds of information does not affect the treatments and patient outcomes, the incompleteness of data reduces the predictive power of ML algorithms and requires significant human and computational resources to account for the missing data and construct reliable computational systems in clinics.…”