Using Big Data Analytics to Advance Precision Radiation Oncology

Data Science Opportunities To Improve Radiotherapy Planning and Clinical Decision Making

Radiotherapy aims to achieve a high tumor control probability while minimizing damage to normal tissues. Personalizing radiotherapy treatments for individual patients, therefore, depends on integrating physical treatment planning with predictive models of tumor control and normal tissue complications. Predictive models could be improved using a wide range of rich data sources, including tumor and normal tissue genomics, radiomics, and dosiomics. Deep learning will drive improvements in classifying normal tissue tolerance, predicting intra-treatment tumor changes, tracking accumulated dose distributions, and quantifying the tumor response to radiotherapy based on imaging. Mechanistic patient-specific computer simulations ('digital twins') could also be used to guide adaptive radiotherapy. Overall, we are entering an era where improved modeling methods will allow the use of newly available data sources to better guide radiotherapy treatments.

Impact of Proton Irradiation Depending on Breast Cancer Subtype in Patient-Derived Cell Lines

Research on different types of ionizing radiation's effects has been ongoing for years, revealing its efficacy in damaging cancer cells. Solid tumors comprise diverse cell types, each being able to respond differently to radiation. This study evaluated the radiobiological response of established (MDA-MB-231 (Triple negative breast cancer, TNBC), MCF-7 (Luminal A)) and patient-derived malignant cell lines, cancer-associated fibroblasts, and skin fibroblasts following proton IRR. All cell line types were irradiated with the proton dose of 2, 4, and 6 Gy. The radiobiological response was assessed using clonogenic assay, γH2AX, and p53 staining. It was noticeable that breast cancer lines of different molecular subtypes displayed no significant variations in their response to proton IRR. In terms of cancer-associated fibroblasts extracted from the tumor tissue, the line derived from a TNBC subtype tumor demonstrated higher resistance to ionizing radiation compared to lines isolated from luminal A tumors. Fibroblasts extracted from patients' skin responded identically to all doses of proton radiation. This study emphasizes that tumor response is not exclusively determined by the elimination of breast cancer cells, but also takes into account tumor microenvironmental variables and skin reactions.

21st century (clinical) decision support in nursing and allied healthcare. Developing a learning health system: a reasoned design of a theoretical framework

In this paper, we present a framework for developing a Learning Health System (LHS) to provide means to a computerized clinical decision support system for allied healthcare and/or nursing professionals. LHSs are well suited to transform healthcare systems in a mission-oriented approach, and is being adopted by an increasing number of countries. Our theoretical framework provides a blueprint for organizing such a transformation with help of evidence based state of the art methodologies and techniques to eventually optimize personalized health and healthcare. Learning via health information technologies using LHS enables users to learn both individually and collectively, and independent of their location. These developments demand healthcare innovations beyond a disease focused orientation since clinical decision making in allied healthcare and nursing is mainly based on aspects of individuals' functioning, wellbeing and (dis)abilities. Developing LHSs depends heavily on intertwined social and technological innovation, and research and development. Crucial factors may be the transformation of the Internet of Things into the Internet of FAIR data & services. However, Electronic Health Record (EHR) data is in up to 80% unstructured including free text narratives and stored in various inaccessible data warehouses. Enabling the use of data as a driver for learning is challenged by interoperability and reusability.To address technical needs, key enabling technologies are suitable to convert relevant health data into machine actionable data and to develop algorithms for computerized decision support. To enable data conversions, existing classification and terminology systems serve as definition providers for natural language processing through (un)supervised learning.To facilitate clinical reasoning and personalized healthcare using LHSs, the development of personomics and functionomics are useful in allied healthcare and nursing. Developing these omics will be determined via text and data mining. This will focus on the relationships between social, psychological, cultural, behavioral and economic determinants, and human functioning.Furthermore, multiparty collaboration is crucial to develop LHSs, and man-machine interaction studies are required to develop a functional design and prototype. During development, validation and maintenance of the LHS continuous attention for challenges like data-drift, ethical, technical and practical implementation difficulties is required.

Prediction of p53 mutation status in rectal cancer patients based on magnetic resonance imaging-based nomogram: a study of machine learning

BACKGROUND

The current study aimed to construct and validate a magnetic resonance imaging (MRI)-based radiomics nomogram to predict tumor protein p53 gene status in rectal cancer patients using machine learning.

METHODS

Clinical and imaging data from 300 rectal cancer patients who underwent radical resections were included in this study, and a total of 166 patients with p53 mutations according to pathology reports were included in these patients. These patients were allocated to the training (n = 210) or validation (n = 90) cohorts (7:3 ratio) according to the examination time. Using the training data set, the radiomic features of primary tumor lesions from T2-weighted images (T2WI) of each patient were analyzed by dimensionality reduction. Multivariate logistic regression was used to screen predictive features, which were combined with a radiomics model to construct a nomogram to predict p53 gene status. The accuracy and reliability of the nomograms were assessed in both training and validation data sets using receiver operating characteristic (ROC) curves.

RESULTS

Using the radiomics model with the training and validation cohorts, the diagnostic efficacies were 0.828 and 0.795, the sensitivities were 0.825 and 0.891, and the specificities were 0.722 and 0.659, respectively. Using the nomogram with the training and validation data sets, the diagnostic efficacies were 0.86 and 0.847, the sensitivities were 0.758 and 0.869, and the specificities were 0.833 and 0.75, respectively.

CONCLUSIONS

The radiomics nomogram based on machine learning was able to predict p53 gene status and facilitate preoperative molecular-based pathological diagnoses.

Quantifying the Growth of Glioblastoma Tumors Using Multimodal MRI Brain Images

Predicting the eventual volume of tumor cells, that might proliferate from a given tumor, can help in cancer early detection and medical procedure planning to prevent their migration to other organs. In this work, a new statistical framework is proposed using Bayesian techniques for detecting the eventual volume of cells expected to proliferate from a glioblastoma (GBM) tumor. Specifically, the tumor region was first extracted using a parallel image segmentation algorithm. Once the tumor region was determined, we were interested in the number of cells that could proliferate from this tumor until its survival time. For this, we constructed the posterior distribution of the tumor cell numbers based on the proposed likelihood function and a certain prior volume. Furthermore, we extended the detection model and conducted a Bayesian regression analysis by incorporating radiomic features to discover those non-tumor cells that remained undetected. The main focus of the study was to develop a time-independent prediction model that could reliably predict the ultimate volume a malignant tumor of the fourth-grade severity could attain and which could also determine if the incorporation of the radiomic properties of the tumor enhanced the chances of no malignant cells remaining undetected.

Identifying requisite learning health system competencies: a scoping review

OBJECTIVES

Learning health systems (LHS) integrate knowledge and practice through cycles of continuous quality improvement and learning to increase healthcare quality. LHS have been conceptualised through multiple frameworks and models. Our aim is to identify and describe the requisite individual competencies (knowledge, skills and attitudes) and system competencies (capacities, characteristics and capabilities) described in existing literature in relation to operationalising LHS.

METHODS

A scoping review was conducted with descriptive and thematic analysis to identify and map competencies of LHS for individuals/patients, health system workers and systems. Articles until April 2020 were included based on a systematic literature search and selection process. Themes were developed using a consensus process until agreement was reached among team members.

RESULTS

Eighty-nine articles were included with most studies conducted in the USA (68 articles). The largest number of publications represented competencies at the system level, followed by health system worker competencies. Themes identified at the individual/patient level were knowledge and skills to understand and share information with an established system and the ability to interact with the technology used to collect data. Themes at the health system worker level were skills in evidence-based practice, leadership and teamwork skills, analytical and technological skills required to use a 'digital ecosystem', data-science knowledge and skill and self-reflective capacity. Researchers embedded within LHS require a specific set of competencies. Themes identified at the system level were data, infrastructure and standardisation; integration of data and workflow; and culture and climate supporting ongoing learning.

CONCLUSION

The identified individual stakeholder competencies within LHS and the system capabilities of LHS provide a solid base for the further development and evaluation of LHS. International collaboration for stimulating LHS will assist in further establishing the knowledge base for LHS.

Concepts and Trends n Autonomy for Robot-Assisted Surgery

Surgical robots have been widely adopted with over 4000 robots being used in practice daily. However, these are telerobots that are fully controlled by skilled human surgeons. Introducing "surgeon-assist"-some forms of autonomy-has the potential to reduce tedium and increase consistency, analogous to driver-assist functions for lanekeeping, cruise control, and parking. This article examines the scientific and technical backgrounds of robotic autonomy in surgery and some ethical, social, and legal implications. We describe several autonomous surgical tasks that have been automated in laboratory settings, and research concepts and trends.

High-dimensional role of AI and machine learning in cancer research

The role of Artificial Intelligence and Machine Learning in cancer research offers several advantages, primarily scaling up the information processing and increasing the accuracy of the clinical decision-making. The key enabling tools currently in use in Precision, Digital and Translational Medicine, here named as 'Intelligent Systems' (IS), leverage unprecedented data volumes and aim to model their underlying heterogeneous influences and variables correlated with patients' outcomes. As functionality and performance of IS are associated with complex diagnosis and therapy decisions, a rich spectrum of patterns and features detected in high-dimensional data may be critical for inference purposes. Many challenges are also present in such discovery task. First, the generation of interpretable model results from a mix of structured and unstructured input information. Second, the design, and implementation of automated clinical decision processes for drawing disease trajectories and patient profiles. Ultimately, the clinical impacts depend on the data effectively subjected to steps such as harmonisation, integration, validation, etc. The aim of this work is to discuss the transformative value of IS applied to multimodal data acquired through various interrelated cancer domains (high-throughput genomics, experimental biology, medical image processing, radiomics, patient electronic records, etc.).

Enhancing the impact of Artificial Intelligence in Medicine: A joint AIFM-INFN Italian initiative for a dedicated cloud-based computing infrastructure

Artificial Intelligence (AI) techniques have been implemented in the field of Medical Imaging for more than forty years. Medical Physicists, Clinicians and Computer Scientists have been collaborating since the beginning to realize software solutions to enhance the informative content of medical images, including AI-based support systems for image interpretation. Despite the recent massive progress in this field due to the current emphasis on Radiomics, Machine Learning and Deep Learning, there are still some barriers to overcome before these tools are fully integrated into the clinical workflows to finally enable a precision medicine approach to patients' care. Nowadays, as Medical Imaging has entered the Big Data era, innovative solutions to efficiently deal with huge amounts of data and to exploit large and distributed computing resources are urgently needed. In the framework of a collaboration agreement between the Italian Association of Medical Physicists (AIFM) and the National Institute for Nuclear Physics (INFN), we propose a model of an intensive computing infrastructure, especially suited for training AI models, equipped with secure storage systems, compliant with data protection regulation, which will accelerate the development and extensive validation of AI-based solutions in the Medical Imaging field of research. This solution can be developed and made operational by Physicists and Computer Scientists working on complementary fields of research in Physics, such as High Energy Physics and Medical Physics, who have all the necessary skills to tailor the AI-technology to the needs of the Medical Imaging community and to shorten the pathway towards the clinical applicability of AI-based decision support systems.

Learning healthcare systems and rapid learning in radiation oncology: Where are we and where are we going?

Learning health systems and rapid-learning are well developed at the conceptual level. The promise of rapidly generating and applying evidence where conventional clinical trials would not usually be practical is attractive in principle. The connectivity of modern digital healthcare information systems and the increasing volumes of data accrued through patients' care pathways offer an ideal platform for the concepts. This is particularly true in radiotherapy where modern treatment planning and image guidance offers a precise digital record of the treatment planned and delivered. The vision is of real-world data, accrued by patients during their routine care, being used to drive programmes of continuous clinical improvement as part of standard practice. This vision, however, is not yet a reality in radiotherapy departments. In this article we review the literature to explore why this is not the case, identify barriers to its implementation, and suggest how wider clinical application might be achieved.

Personalization in Modern Radiation Oncology: Methods, Results and Pitfalls. Personalized Interventions and Breast Cancer

Breast cancer, the most frequent malignancy in women worldwide, is a heterogeneous group of diseases, characterized by distinct molecular aberrations. In precision medicine, radiation oncology for breast cancer aims at tailoring treatment according to tumor biology and each patient’s clinical features and genetics. Although systemic therapies are personalized according to molecular sub-type [i.e. endocrine therapy for receptor-positive disease and anti-human epidermal growth factor receptor 2 (HER2) therapy for HER2-positive disease] and multi-gene assays, personalized radiation therapy has yet to be adopted in the clinical setting. Currently, attempts are being made to identify prognostic and/or predictive factors, biomarkers, signatures that could lead to personalized treatment in order to select appropriate patients who might, or might not, benefit from radiation therapy or whose radiation therapy might be escalated or de-escalated in dosages and volumes. This overview focuses on what has been achieved to date in personalized post-operative radiation therapy and individual patient radiosensitivity assessments by means of tumor sub-types and genetics.

Role of radiation oncology in modern multidisciplinary cancer treatment

Cancer care is moving from a disease‐focused management toward a patient‐centered tailored approach. Multidisciplinary management that aims to define individual, optimal treatment strategies through shared decision making between healthcare professionals and patient is a fundamental aspect of high‐quality cancer care and often includes radiation oncology. Advances in technology and radiobiological research allow to deliver ever more tailored radiation treatments in an ever easier and faster way, thus improving the efficacy, safety, and accessibility of radiation therapy. While these changes are improving quality of cancer care, they are also enormously increasing complexity of decision making, thus challenging the ability to deliver quality affordable cancer care. In this review, we provide an updated outline of the role of radiation oncology in the modern multidisciplinary treatment of cancer. Particularly, we focus on the way some developments in key areas of cancer management are challenging multidisciplinary cancer care in the different clinical settings of early, locally advanced, and metastatic disease, thus highlighting some priority areas of research.

"Après Mois, Le Déluge": Preparing for the Coming Data Flood in the MRI-Guided Radiotherapy Era

Magnetic resonance imaging provides a sea of quantitative and semi-quantitative data. While radiation oncologists already navigate a pool of clinical (semantic) and imaging data, the tide will swell with the advent of hybrid MRI/linear accelerator devices and increasing interest in MRI-guided radiotherapy (MRIgRT), including adaptive MRIgRT. The variety of MR sequences (of greater complexity than the single parameter Hounsfield unit of CT scanning routinely used in radiotherapy), the workflow of adaptive fractionation, and the sheer quantity of daily images acquired are challenges for scaling this technology. Biomedical informatics, which is the science of information in biomedicine, can provide helpful insights for this looming transition. Funneling MRIgRT data into clinically meaningful information streams requires committing to the flow of inter-institutional data accessibility and interoperability initiatives, standardizing MRIgRT dosimetry methods, streamlining MR linear accelerator workflow, and standardizing MRI acquisition and post-processing. This review will attempt to conceptually ford these topics using clinical informatics approaches as a theoretical bridge.

Predicting acute radiation induced xerostomia in head and neck Cancer using MR and CT Radiomics of parotid and submandibular glands

PURPOSE

To analyze baseline CT/MR-based image features of salivary glands to predict radiation-induced xerostomia 3-months after head-and-neck cancer (HNC) radiotherapy.

METHODS

A retrospective analysis was performed on 266 HNC patients who were treated using radiotherapy at our institution between 2009 and 2018. CT and T1 post-contrast MR images along with NCI-CTCAE xerostomia grade (3-month follow-up) were prospectively collected at our institution. CT and MR images were registered on which parotid/submandibular glands were contoured. Image features were extracted for ipsilateral/contralateral parotid and submandibular glands relative to the location of the primary tumor. Dose-volume-histogram (DVH) parameters were also acquired. Features were pre-selected based on Spearman correlation before modelling by examining the correlation with xerostomia (p < 0.05). A shrinkage regression analysis of the pre-selected features was performed using LASSO. The internal validity of the variable selection was estimated by repeating the entire variable selection procedure using a leave-one-out-cross-validation. The most frequently selected variables were considered in the final model. A generalized linear regression with repeated ten-fold cross-validation was developed to predict radiation-induced xerostomia at 3-months after radiotherapy. This model was tested in an independent dataset (n = 50) of patients who were treated at the same institution in 2017-2018. We compared the prediction performances under eight conditions (DVH-only, CT-only, MR-only, CT + MR, DVH + CT, DVH + CT + MR, Clinical+CT + MR, and Clinical+DVH + CT + MR) using the area under the receiver operating characteristic curve (ROC-AUC).

RESULTS

Among extracted features, 7 CT, 5 MR, and 2 DVH features were selected. The internal cohort (n = 216) ROC-AUC values for DVH, CT, MR, and Clinical+DVH + CT + MR features were 0.73 ± 0.01, 0.69 ± 0.01, 0.70 ± 0.01, and 0.79 ± 0.01, respectively. The validation cohort (n = 50) ROC-AUC values for DVH, CT, MR, and Clinical+DVH + CT + MR features were 0.63, 0.57, 0.66, and 0.68, respectively. The DVH-ROC was not significantly different than the CT-ROC (p = 0.8) or MR-ROC (p = 0.4). However, the CT + MR-ROC was significantly different than the CT-ROC (p = 0.03), but not the Clinical+DVH + CT + MR model (p = 0.5).

CONCLUSION

Our results suggest that baseline CT and MR image features may reflect baseline salivary gland function and potential risk for radiation injury. The integration of baseline image features into prediction models has the potential to improve xerostomia risk stratification with the ultimate goal of truly personalized HNC radiotherapy.

Decision Support Systems in Prostate Cancer Treatment: An Overview

Background

A multifactorial decision support system (mDSS) is a tool designed to improve the clinical decision-making process, while using clinical inputs for an individual patient to generate case-specific advice. The study provides an overview of the literature to analyze current available mDSS focused on prostate cancer (PCa), in order to better understand the availability of decision support tools as well as where the current literature is lacking.

Methods

We performed a MEDLINE literature search in July 2018. We divided the included studies into different sections: diagnostic, which aids in detection or staging of PCa; treatment, supporting the decision between treatment modalities; and patient, which focusses on informing the patient. We manually screened and excluded studies that did not contain an mDSS concerning prostate cancer and study proposals.

Results

Our search resulted in twelve diagnostic mDSS; six treatment mDSS; two patient mDSS; and eight papers that could improve mDSS.

Conclusions

Diagnosis mDSS is well represented in the literature as well as treatment mDSS considering external-beam radiotherapy; however, there is a lack of mDSS for other treatment modalities. The development of patient decision aids is a new field of research, and few successes have been made for PCa patients. These tools can improve personalized medicine but need to overcome a number of difficulties to be successful and require more research.

Radiomics in Glioblastoma: Current Status and Challenges Facing Clinical Implementation

Radiomics analysis has had remarkable progress along with advances in medical imaging, most notability in central nervous system malignancies. Radiomics refers to the extraction of a large number of quantitative features that describe the intensity, texture and geometrical characteristics attributed to the tumor radiographic data. These features have been used to build predictive models for diagnosis, prognosis, and therapeutic response. Such models are being combined with clinical, biological, genetics and proteomic features to enhance reproducibility. Broadly, the four steps necessary for radiomic analysis are: (1) image acquisition, (2) segmentation or labeling, (3) feature extraction, and (4) statistical analysis. Major methodological challenges remain prior to clinical implementation. Essential steps include: adoption of an optimized standard imaging process, establishing a common criterion for performing segmentation, fully automated extraction of radiomic features without redundancy, and robust statistical modeling validated in the prospective setting. This review walks through these steps in detail, as it pertains to high grade gliomas. The impact on precision medicine will be discussed, as well as the challenges facing clinical implementation of radiomic in the current management of glioblastoma.

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

For nearly 2 decades, adaptive radiation therapy (ART) has been proposed as a method to account for changes in head and neck tumor and normal tissue to enhance therapeutic ratios. While technical advances in imaging, planning and delivery have allowed greater capacity for ART delivery, and a series of dosimetric explorations have consistently shown capacity for improvement, there remains a paucity of clinical trials demonstrating the utility of ART. Furthermore, while ad hoc implementation of head and neck ART is reported, systematic full-scale head and neck ART remains an as yet unreached reality. To some degree, this lack of scalability may be related to not only the complexity of ART, but also variability in the nomenclature and descriptions of what is encompassed by ART. Consequently, we present an overview of the history, current status, and recommendations for the future of ART, with an eye toward improving the clarity and description of head and neck ART for interested clinicians, noting practical considerations for implementation of an ART program or clinical trial. Process level considerations for ART are noted, reminding the reader that, paraphrasing the writer Elbert Hubbard, "Art is not a thing, it is a way."

Dose/Volume histogram patterns in Salivary Gland subvolumes influence xerostomia injury and recovery

Xerostomia is a common consequence of radiotherapy in head and neck cancer. The objective was to compare the regional radiation dose distribution in patients that developed xerostomia within 6 months of radiotherapy and those recovered from xerostomia within 18 months post-radiotherapy. We developed a feature generation pipeline to extract dose volume histogram features from geometrically defined ipsilateral/contralateral parotid glands, submandibular glands, and oral cavity surrogates for each patient. Permutation tests with multiple comparisons were performed to assess the dose difference between injury vs. non-injury and recovery vs. non-recovery. Ridge logistic regression models were applied to predict injury and recovery using clinical features along with dose features (D10-D90) of the subvolumes extracted from oral cavity and salivary gland contours + 3 mm peripheral shell. Model performances were assessed by the area under the receiver operating characteristic curve (AUC) using nested cross-validation. We found that different regional dose/volume metrics patterns exist for injury vs. recovery. Compared to injury, recovery has increased importance to the subvolumes receiving lower dose. Within the subvolumes, injury tends to have increased importance towards D10 from D90. This suggests that different threshold for xerostomia injury and recovery. Injury is induced by the subvolumes receiving higher dose, and the ability to recover can be preserved by further reducing the dose to subvolumes receiving lower dose.

Needs and Challenges for Radiation Oncology in the Era of Precision Medicine

Modern medicine, including the care of the cancer patient, has significantly advanced, with the evidence-based medicine paradigm serving to guide clinical care decisions. Yet we now also recognize the tremendous heterogeneity not only of disease states but of the patient and his or her environment as it influences treatment outcomes and toxicities. These reasons and many others have led to a reevaluation of the generalizability of randomized trials and growing interest in accounting for this heterogeneity under the rubric of precision medicine as it relates to personalizing clinical care predictions, decisions, and therapy for the disease state. For the cancer patient treated with radiation therapy, characterizing the spatial treatment heterogeneity has been a fundamental tenet of routine clinical care facilitated by established database and imaging platforms. Leveraging these platforms to further characterize and collate all clinically relevant sources of heterogeneity that affect the longitudinal health outcomes of the irradiated cancer patient provides an opportunity to generate a critical informatics infrastructure on which precision radiation therapy may be realized. In doing so, data science-driven insight discoveries, personalized clinical decisions, and the potential to accelerate translational efforts may be realized ideally within a network of institutions with locally developed yet coordinated informatics infrastructures. The path toward realizing these goals has many needs and challenges, which we summarize, with many still to be realized and understood. Early efforts by our group have identified the feasibility of this approach using routine clinical data sets and offer promise that this transformation can be successfully realized in radiation oncology.

The Needs and Benefits of Continuous Model Updates on the Accuracy of RT-Induced Toxicity Prediction Models Within a Learning Health System

PURPOSE

Clinical data collection and development of outcome prediction models by machine learning can form the foundation for a learning health system offering precision radiation therapy. However, changes in clinical practice over time can affect the measures and patient outcomes and, hence, the collected data. We hypothesize that regular prediction model updates and continuous prospective data collection are important to prevent the degradation of a model's predication accuracy.

METHODS AND MATERIALS

Clinical and dosimetric data from head and neck patients receiving intensity modulated radiation therapy from 2008 to 2015 were prospectively collected as a routine clinical workflow and anonymized for this analysis. Prediction models for grade ≥2 xerostomia at 3 to 6 months of follow-up were developed by bivariate logistic regression using the dose-volume histogram of parotid and submandibular glands. A baseline prediction model was developed with a training data set from 2008 to 2009. The selected predictor variables and coefficients were updated by 4 different model updating methods. (A) The prediction model was updated by using only recent 2-year data and applied to patients in the following test year. (B) The model was updated by increasing the training data set yearly. (C) The model was updated by increasing the training data set on the condition that the area under the curve (AUC) of the recent test year was less than 0.6. (D) The model was not updated. The AUC of the test data set was compared among the 4 model updating methods.

RESULTS

Dose to parotid and submandibular glands and grade of xerostomia showed decreasing trends over the years (2008-2015, 297 patients; P < .001). The AUC of predicting grade ≥2 xerostomia for the initial training data set (2008-2009, 41 patients) was 0.6196. The AUC for the test data set (2010-2015, 256 patients) decreased to 0.5284 when the initial model was not updated (D). However, the AUC was significantly improved by model updates (A: 0.6164; B: 0.6084; P < .05). When the model was conditionally updated, the AUC was 0.6072 (C).

CONCLUSIONS

Our preliminary results demonstrate that updating prediction models with prospective data collection is effective for maintaining the performance of xerostomia prediction. This suggests that a machine learning framework can handle the dynamic changes in a radiation oncology clinical practice and may be an important component for the construction of a learning health system.