Treatment planning is an essential step of the radiotherapy workflow. It has become more sophisticated over the past couple of decades with the help of computer science, enabling planners to design highly complex radiotherapy plans to minimize the normal tissue damage while persevering sufficient tumor control. As a result, treatment planning has become more labor intensive, requiring hours or even days of planner effort to optimize an individual patient case in a trial-and-error fashion. More recently, artificial intelligence has been utilized to automate and improve various aspects of medical science. For radiotherapy treatment planning, many algorithms have been developed to better support planners. These algorithms focus on automating the planning process and/or optimizing dosimetric trade-offs, and they have already made great impact on improving treatment planning efficiency and plan quality consistency. In this review, the smart planning tools in current clinical use are summarized in 3 main categories: automated rule implementation and reasoning, modeling of prior knowledge in clinical practice, and multicriteria optimization. Novel artificial intelligence–based treatment planning applications, such as deep learning–based algorithms and emerging research directions, are also reviewed. Finally, the challenges of artificial intelligence–based treatment planning are discussed for future works.

An 18-patient study demonstrated significant correlations between 4D-CT ventilation and PFT measurements as well as SPECT ventilation, providing evidence toward the validation of 4D-CT ventilation imaging.

BackgroundRadiotherapy is increasingly used to treat oligometastatic patients. We sought to identify prognostic criteria in oligometastatic patients undergoing definitive hypofractionated image-guided radiotherapy (HIGRT).MethodsExclusively extracranial oligometastatic patients treated with HIGRT were pooled. Characteristics including age, sex, primary tumor type, interval to metastatic diagnosis, number of treated metastases and organs, metastatic site, prior systemic therapy for primary tumor treatment, prior definitive metastasis-directed therapy, and systemic therapy for metastasis associated with overall survival (OS), progression-free survival (PFS), and treated metastasis control (TMC) were assessed by the Cox proportional hazards method. Recursive partitioning analysis (RPA) identified prognostic risk strata for OS and PFS based on pretreatment factors.Results361 patients were included. Primary tumors included non-small cell lung (17%), colorectal (19%), and breast cancer (16%). Three-year OS was 56%, PFS was 24%, and TMC was 72%. On multivariate analysis, primary tumor, interval to metastases, treated metastases number, and mediastinal/hilar lymph node, liver, or adrenal metastases were associated with OS. Primary tumor site, involved organ number, liver metastasis, and prior primary disease chemotherapy were associated with PFS.OS RPA identified five classes: class 1: all breast, kidney, or prostate cancer patients (BKP) (3-year OS 75%, 95% CI 66–85%); class 2: patients without BKP with disease-free interval of 75+ months (3-year OS 85%, 95% CI 67–100%); class 3: patients without BKP, shorter disease-free interval, ≤ two metastases, and age < 62 (3-year OS 55%, 95% CI 48–64%); class 4: patients without BKP, shorter disease-free interval, ≥ three metastases, and age < 62 (3-year OS 38%, 95% CI 24–60%); class 5: all others (3-year OS 13%, 95% CI 5–35%). Higher biologically effective dose (BED) (p < 0.01) was associated with OS.ConclusionsWe identified clinical factors defining oligometastatic patients with favorable outcomes, who we hypothesize are most likely to benefit from metastasis-directed therapy.

PURPOSE Patients undergoing outpatient radiotherapy (RT) or chemoradiation (CRT) frequently require acute care (emergency department evaluation or hospitalization). Machine learning (ML) may guide interventions to reduce this risk. There are limited prospective studies investigating the clinical impact of ML in health care. The objective of this study was to determine whether ML can identify high-risk patients and direct mandatory twice-weekly clinical evaluation to reduce acute care visits during treatment. PATIENTS AND METHODS During this single-institution randomized quality improvement study (ClinicalTrials.gov identifier: NCT04277650 ), 963 outpatient adult courses of RT and CRT started from January 7 to June 30, 2019, were evaluated by an ML algorithm. Among these, 311 courses identified by ML as high risk (> 10% risk of acute care during treatment) were randomized to standard once-weekly clinical evaluation (n = 157) or mandatory twice-weekly evaluation (n = 154). Both arms allowed additional evaluations on the basis of clinician discretion. The primary end point was the rate of acute care visits during RT. Model performance was evaluated using receiver operating characteristic area under the curve (AUC) and decile calibration plots. RESULTS Twice-weekly evaluation reduced rates of acute care during treatment from 22.3% to 12.3% (difference, −10.0%; 95% CI, −18.3 to −1.6; relative risk, 0.556; 95% CI, 0.332 to 0.924; P = .02). Low-risk patients had a 2.7% acute care rate. Model discrimination was good in high- and low-risk patients undergoing standard once-weekly evaluation (AUC, 0.851). CONCLUSION In this prospective randomized study, ML accurately triaged patients undergoing RT and CRT, directing clinical management with reduced acute care rates versus standard of care. This prospective study demonstrates the potential benefit of ML in health care and offers opportunities to enhance care quality and reduce health care costs.

Rationale and Objectives A novel ventilation imaging method based on four-dimensional (4D) computed tomography (CT) has been applied to the field of radiation oncology. Understanding its reproducibility is a prerequisite for clinical applications. The purpose of this study was to quantify the reproducibility of 4D CT ventilation imaging over different days and the same session. Materials and Methods Two ventilation images were created from repeat 4D CT scans acquired over the average time frames of 15 days for 6 lung cancer patients and 5 minutes for another 6 patients. The reproducibility was quantified using the voxel-based Spearman rank correlation coefficients for all lung voxels and Dice similarity coefficients (DSC) for the spatial overlap of segmented high-, moderate-, and low-functional lung volumes. Furthermore, the relationship between the variation in abdominal motion range as a measure of the depth of breathing and variation in ventilation was evaluated using linear regression. Results The voxel-based correlation between the two ventilation images was moderate on average (0.50 ± 0.15). The DSCs were also moderate for the high- (0.60 ± 0.08), moderate- (0.46 ± 0.06), and low-functional lung (0.58 ± 0.09). No patients demonstrated strong correlations. The relationship between the motion range variation and ventilation variation was found to be moderate and significant. Conclusions We investigated the reproducibility of 4D CT ventilation imaging over the time frames of 15 days and 5 minutes and found that it was only moderately reproducible. Respiratory variation during 4D CT scans was found to deteriorate the reproducibility. Improvement of 4D CT imaging is necessary to increase the reproducibility of 4D CT ventilation imaging.

Background and purpose Loco-regionally recurrent head and neck cancer (HNC) in the setting of prior radiotherapy carries significant morbidity and mortality. The role of re-irradiation (re-RT) remains unclear due to toxicity. We determined prognostic factors for loco-regional control (LRC) and formulated a nomogram to help clinicians select re-RT candidates. Material and methods From July 1996 to April 2011, 257 patients with recurrent HNC underwent fractionated re-RT. Median prior dose was 65 Gy and median time between RT was 32.4 months. One hundred fifteen patients (44%) had salvage surgery and 172 (67%) received concurrent chemotherapy. Median re-RT dose was 59.4 Gy and 201 (78%) patients received IMRT. Multivariate Cox proportional hazards were used to identify independent predictors of LRC and a nomogram for 2-year LRC was constructed. Results Median follow-up was 32.6 months. Two-year LRC and overall survival (OS) were 47% and 43%, respectively. Recurrent stage (P = 0.005), non-oral cavity subsite (P < 0.001), absent organ dysfunction (P < 0.001), salvage surgery (P < 0.001), and dose >50 Gy (P = 0.006) were independently associated with improved LRC. We generated a nomogram with concordance index of 0.68. Conclusion Re-RT can be curative, and our nomogram can help determine a priori which patients may benefit.

Purpose Patients undergoing radiotherapy (RT) or chemoradiotherapy (CRT) may require emergency department evaluation or hospitalization. Early identification may direct preventative supportive care, improving outcomes and reducing health care costs. We developed and evaluated a machine learning (ML) approach to predict these events. Methods A total of 8,134 outpatient courses of RT and CRT from a single institution from 2013 to 2016 were identified. Extensive pretreatment data were programmatically extracted and processed from the electronic health record (EHR). Training and internal validation cohorts were randomly generated (3:1 ratio). Gradient tree boosting (GTB), random forest, support vector machine, and least absolute shrinkage and selection operator logistic regression approaches were trained and internally validated based on area under receiver operating characteristic (AUROC) curve. The most predictive ML approach was also evaluated using only disease- and treatment-related factors to assess predictive gain of extensive EHR data. Results All methods had high predictive accuracy, particularly GTB (validation AUROC, 0.798). Extensive EHR data beyond disease and treatment information improved accuracy (delta AUROC, 0.056). A Youden-based cutoff corresponded to validation sensitivity of 81.0% (175 of 216 courses with events) and specificity of 67.3% (1,218 of 1811 courses without events). Interpretability is an important advantage of GTB. Variable importance identified top predictive factors, including treatment (planned RT and systemic therapy), pretreatment encounters (emergency department visits and admissions in the year before treatment), vital signs (weight loss and pain score in the year before treatment), and laboratory values (albumin level at weeks before treatment). Conclusion ML predicts emergency visits and hospitalization during cancer therapy. Incorporating predictions into clinical care algorithms may help direct personalized supportive care, improve quality of care, and reduce costs. A prospective trial investigating ML-assisted direction of increased clinical assessments during RT is planned.

The purpose of this work was to investigate the potential relationship between radiomic features extracted from pre-treatment x-ray CT images and clinical outcomes following stereotactic body radiation therapy (SBRT) for non-small-cell lung cancer (NSCLC). Seventy patients who received SBRT for stage-1 NSCLC were retrospectively identified. The tumor was contoured on pre-treatment free-breathing CT images, from which 43 quantitative radiomic features were extracted to collectively capture tumor morphology, intensity, fine-texture, and coarse-texture. Treatment failure was defined based on cancer recurrence, local cancer recurrence, and non-local cancer recurrence following SBRT. The univariate association between each radiomic feature and each clinical endpoint was analyzed using Welch's t-test, and p-values were corrected for multiple hypothesis testing. Multivariate associations were based on regularized logistic regression with a singular value decomposition to reduce the dimensionality of the radiomics data. Two features demonstrated a statistically significant association with local failure: Homogeneity2 (p = 0.022) and Long-Run-High-Gray-Level-Emphasis (p = 0.048). These results indicate that relatively dense tumors with a homogenous coarse texture might be linked to higher rates of local recurrence. Multivariable logistic regression models produced maximum AUC values of 0.72 ± 0.04, 0.83 ± 0.03, and 0.60 ± 0.04, for the recurrence, local recurrence, and non-local recurrence endpoints, respectively. The CT-based radiomic features used in this study may be more associated with local failure than non-local failure following SBRT for stage I NSCLC. This finding is supported by both univariate and multivariate analyses.