Support Vector Machine-Based Prediction of Local Tumor Control After Stereotactic Body Radiation Therapy for Early-Stage Non-Small Cell Lung Cancer

Artificial Neural Network Prediction of Mortality in Cancer Patients Presenting for Radiation Therapy at a Multisite Institution

INTRODUCTION

For many decades, the management of cancer has utilized radiation therapy, which continues to evolve with technology to improve patient outcomes. However, despite the standardization of treatment plans and the establishment of best clinical practices based on prospective, randomized trials and adherence to National Comprehensive Cancer Network (NCCN) guidelines, the outcomes from radiation therapy are highly variable and dependent on a number of factors, including patient demographics, tumor characteristics/histology, and treatment parameters. In this study, we attempt to use available patient data and treatment parameters at the time of radiation therapy to predict future outcomes using artificial intelligence (AI).

METHODS

Six thousand five hundred ninety-five cases of patients who completed radiation treatment were selected retrospectively and used to train artificial neural networks (ANNs) and baseline models (i.e., logistic regression, random forest, support vector machines [SVMs], gradient boosting [XGBoost]) for binary classification of mortality at multiple time points ranging from six months to five years post-treatment. A hyperparameter grid search was used to identify the optimal network architecture for each time point, using sensitivity as the primary outcome metric.

RESULTS

The median age was 75 years (range: 2-102 years). There were 63.8% females and 36.1% males. The results indicate that ANNs were able to successfully perform binary mortality prediction with an accuracy greater than random chance and greater sensitivity than baseline models used. The best-performing algorithm was the ANN, which achieved a sensitivity of 83.00% ± 4.89% for five-year mortality.

CONCLUSION

The neural network was able to achieve higher sensitivity than Logistic Regression, SVM Random Forest, and XGBoost across all output target variables, demonstrating the utility of a neural network model for mortality prediction on the provided dataset.

Deep Learning-Guided Dosimetry for Mitigating Local Failure of Patients With Non-Small Cell Lung Cancer Receiving Stereotactic Body Radiation Therapy

PURPOSE

Non-small cell lung cancer (NSCLC) stereotactic body radiation therapy with 50 Gy/5 fractions is sometimes considered controversial, as the nominal biologically effective dose (BED) of 100 Gy is felt by some to be insufficient for long-term local control of some lesions. In this study, we analyzed such patients using explainable deep learning techniques and consequently proposed appropriate treatment planning criteria. These novel criteria could help planners achieve optimized treatment plans for maximal local control.

METHODS AND MATERIALS

A total of 535 patients treated with 50 Gy/5 fractions were used to develop a novel deep learning local response model. A multimodality approach, incorporating computed tomography images, 3-dimensional dose distribution, and patient demographics, combined with a discrete-time survival model, was applied to predict time to failure and the probability of local control. Subsequently, an integrated gradient-weighted class activation mapping method was used to identify the most significant dose-volume metrics predictive of local failure and their optimal cut-points.

RESULTS

The model was cross-validated, showing an acceptable performance (c-index: 0.72, 95% CI, 0.68-0.75); the testing c-index was 0.69. The model's spatial attention was concentrated mostly in the tumors' periphery (planning target volume [PTV] - internal gross target volume [IGTV]) region. Statistically significant dose-volume metrics in improved local control were BED Dnear-min ≥ 103.8 Gy in IGTV (hazard ratio [HR], 0.31; 95% CI, 015-0.63), V104 ≥ 98% in IGTV (HR, 0.30; 95% CI, 0.15-0.60), gEUD ≥ 103.8 Gy in PTV-IGTV (HR, 0.25; 95% CI, 0.12-0.50), and Dmean ≥ 104.5 Gy in PTV-IGTV (HR, 0.25; 95% CI, 0.12-0.51).

CONCLUSIONS

Deep learning-identified dose-volume metrics have shown significant prognostic power (log-rank, P = .003) and could be used as additional actionable criteria for treatment planning in NSCLC stereotactic body radiation therapy patients receiving 50 Gy in 5 fractions. Although our data do not confirm or refute that a significantly higher BED for the prescription dose is necessary for tumor control in NSCLC, it might be clinically effective to escalate the nominal prescribed dose from BED 100 to 105 Gy.

Machine-learning prediction of treatment response to stereotactic body radiation therapy in oligometastatic gynecological cancer: A multi-institutional study

BACKGROUND AND PURPOSE

We aimed to develop and validate different machine-learning (ML) prediction models for the complete response of oligometastatic gynecological cancer after SBRT.

MATERIAL AND METHODS

One hundred fifty-seven patients with 272 lesions from 14 different institutions and treated with SBRT with radical intent were included. Thirteen datasets including 222 lesions were combined for model training and internal validation purposes, with an 80:20 ratio. The external testing dataset was selected as the fourteenth Institution with 50 lesions. Lesions that achieved complete response (CR) were defined as responders. Prognostic clinical and dosimetric variables were selected using the LASSO algorithm. Six supervised ML models, including logistic regression (LR), classification and regression tree analysis (CART) and support vector machine (SVM) using four different kernels, were trained and tested to predict the complete response of uterine lesions after SBRT. The performance of models was assessed by receiver operating characteristic curves (ROC), area under the curve (AUC) and calibration curves. An explainable approach based on SHapley Additive exPlanations (SHAP) method was deployed to generate individual explanations of the model's decisions.

RESULTS

63.6% of lesions had a complete response and were used as ground truth for the supervised models. LASSO strongly associated complete response with three variables, namely the lesion volume (PTV), the type of lesions (lymph-nodal versus parenchymal), and the biological effective dose (BED10), that were used as input for ML modeling. In the training set, the AUCs for complete response were 0.751 (95% CI: 0.716-0.786), 0.766 (95% CI: 0.729-0.802) and 0.800 (95% CI: 0.742-0.857) for the LR, CART and SVM with a radial basis function kernel, respectively. These models achieve AUC values of 0.727 (95% CI: 0.669-0.795), 0.734 (95% CI: 0.649-0.815) and 0.771 (95% CI: 0.717-0.824) in the external testing set, demonstrating excellent generalizability.

CONCLUSION

ML models enable a reliable prediction of the treatment response of oligometastatic lesions receiving SBRT. This approach may assist radiation oncologists to tailor more individualized treatment plans for oligometastatic patients.

Artificial Intelligence in Lung Cancer Imaging: Unfolding the Future

Lung cancer is one of the malignancies with higher morbidity and mortality. Imaging plays an essential role in each phase of lung cancer management, from detection to assessment of response to treatment. The development of imaging-based artificial intelligence (AI) models has the potential to play a key role in early detection and customized treatment planning. Computer-aided detection of lung nodules in screening programs has revolutionized the early detection of the disease. Moreover, the possibility to use AI approaches to identify patients at risk of developing lung cancer during their life can help a more targeted screening program. The combination of imaging features and clinical and laboratory data through AI models is giving promising results in the prediction of patients’ outcomes, response to specific therapies, and risk for toxic reaction development. In this review, we provide an overview of the main imaging AI-based tools in lung cancer imaging, including automated lesion detection, characterization, segmentation, prediction of outcome, and treatment response to provide radiologists and clinicians with the foundation for these applications in a clinical scenario.

Applications of artificial intelligence in stereotactic body radiation therapy

This topical review focuses on the applications of artificial intelligence (AI) tools to stereotactic body radiation therapy (SBRT). The high dose per fraction and the limited number of fractions in SBRT require stricter accuracy than standard radiation therapy. The intent of this review is to describe the development and evaluate the possible benefit of AI tools integration into the radiation oncology workflow for SBRT automation. The selected papers were subdivided into four sections, representative of the whole radiotherapy process: ‘AI in SBRT target and organs at risk contouring’, ‘AI in SBRT planning’, ‘AI during the SBRT delivery’, and ‘AI for outcome prediction after SBRT’. Each section summarises the challenges, as well as limits and needs for improvement to achieve better integration of AI tools in the clinical workflow.

A combined predictive model based on radiomics features and clinical factors for disease progression in early-stage non-small cell lung cancer treated with stereotactic ablative radiotherapy

Purpose

To accurately assess disease progression after Stereotactic Ablative Radiotherapy (SABR) of early-stage Non-Small Cell Lung Cancer (NSCLC), a combined predictive model based on pre-treatment CT radiomics features and clinical factors was established.

Methods

This study retrospectively analyzed the data of 96 patients with early-stage NSCLC treated with SABR. Clinical factors included general information (e.g. gender, age, KPS, Charlson score, lung function, smoking status), pre-treatment lesion status (e.g. diameter, location, pathological type, T stage), radiation parameters (biological effective dose, BED), the type of peritumoral radiation-induced lung injury (RILI). Independent risk factors were screened by logistic regression analysis. Radiomics features were extracted from pre-treatment CT. The minimum Redundancy Maximum Relevance (mRMR) and the Least Absolute Shrinkage and Selection Operator (LASSO) were adopted for the dimensionality reduction and feature selection. According to the weight coefficient of the features, the Radscore was calculated, and the radiomics model was constructed. Multiple logistic regression analysis was applied to establish the combined model based on radiomics features and clinical factors. Receiver Operating Characteristic (ROC) curve, DeLong test, Hosmer-Lemeshow test, and Decision Curve Analysis (DCA) were used to evaluate the model’s diagnostic efficiency and clinical practicability.

Results

With the median follow-up of 59.1 months, 29 patients developed progression and 67 remained good controlled within two years. Among the clinical factors, the type of peritumoral RILI was the only independent risk factor for progression (P< 0.05). Eleven features were selected from 1781 features to construct a radiomics model. For predicting disease progression after SABR, the Area Under the Curve (AUC) of training and validation cohorts in the radiomics model was 0.88 (95%CI 0.80-0.96) and 0.80 (95%CI 0.62-0.98), and AUC of training and validation cohorts in the combined model were 0.88 (95%CI 0.81-0.96) and 0.81 (95%CI 0.62-0.99). Both the radiomics and the combined models have good prediction efficiency in the training and validation cohorts. Still, DeLong test shows that there is no difference between them.

Conclusions

Compared with the clinical model, the radiomics model and the combined model can better predict the disease progression of early-stage NSCLC after SABR, which might contribute to individualized follow-up plans and treatment strategies.

Predictive performance of different NTCP techniques for radiation-induced esophagitis in NSCLC patients receiving proton radiotherapy

This study aimed to compare the predictive performance of different modeling methods in developing normal tissue complication probability (NTCP) models for predicting radiation-induced esophagitis (RE) in non–small cell lung cancer (NSCLC) patients receiving proton radiotherapy. The dataset was composed of 328 NSCLC patients receiving passive-scattering proton therapy and 41.6% of the patients experienced ≥ grade 2 RE. Five modeling methods were used to build NTCP models: standard Lyman–Kutcher–Burman (sLKB), generalized LKB (gLKB), multivariable logistic regression using two variable selection procedures-stepwise forward selection (Stepwise-MLR), and least absolute shrinkage and selection operator (LASSO-MLR), and support vector machines (SVM). Predictive performance was internally validated by a bootstrap approach for each modeling method. The overall performance, discriminative ability, and calibration were assessed using the Negelkerke R², area under the receiver operator curve (AUC), and Hosmer–Lemeshow test, respectively. The LASSO-MLR model showed the best discriminative ability with an AUC value of 0.799 (95% confidence interval (CI): 0.763–0.854), and the best overall performance with a Negelkerke R² value of 0.332 (95% CI: 0.266–0.486). Both of the optimism-corrected Negelkerke R² values of the SVM and sLKB models were 0.301. The optimism-corrected AUC of the gLKB model (0.796) was higher than that of the SVM model (0.784). The sLKB model had the smallest optimism in the model variation and discriminative ability. In the context of classification and probability estimation for predicting the NTCP for radiation-induced esophagitis, the MLR model developed with LASSO provided the best predictive results. The simplest LKB modeling had similar or even better predictive performance than the most complex SVM modeling, and it was least likely to overfit the training data. The advanced machine learning approach might have limited applicability in clinical settings with a relatively small amount of data.

PrognosiT: Pathway/gene set-based tumour volume prediction using multiple kernel learning

Background

Identification of molecular mechanisms that determine tumour progression in cancer patients is a prerequisite for developing new disease treatment guidelines. Even though the predictive performance of current machine learning models is promising, extracting significant and meaningful knowledge from the data simultaneously during the learning process is a difficult task considering the high-dimensional and highly correlated nature of genomic datasets. Thus, there is a need for models that not only predict tumour volume from gene expression data of patients but also use prior information coming from pathway/gene sets during the learning process, to distinguish molecular mechanisms which play crucial role in tumour progression and therefore, disease prognosis.

Results

In this study, instead of initially choosing several pathways/gene sets from an available set and training a model on this previously chosen subset of genomic features, we built a novel machine learning algorithm, PrognosiT, that accomplishes both tasks together. We tested our algorithm on thyroid carcinoma patients using gene expression profiles and cancer-specific pathways/gene sets. Predictive performance of our novel multiple kernel learning algorithm (PrognosiT) was comparable or even better than random forest (RF) and support vector regression (SVR). It is also notable that, to predict tumour volume, PrognosiT used gene expression features less than one-tenth of what RF and SVR algorithms used.

Conclusions

PrognosiT was able to obtain comparable or even better predictive performance than SVR and RF. Moreover, we demonstrated that during the learning process, our algorithm managed to extract relevant and meaningful pathway/gene sets information related to the studied cancer type, which provides insights about its progression and aggressiveness. We also compared gene expressions of the selected genes by our algorithm in tumour and normal tissues, and we then discussed up- and down-regulated genes selected by our algorithm while learning, which could be beneficial for determining new biomarkers.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12859-021-04460-6.

Support Vector Machine Model Predicts Dose for Organs at Risk in High-Dose Rate Brachytherapy of Cervical Cancer

Introduction

This study aimed to establish a support vector machine (SVM) model to predict the dose for organs at risk (OARs) in intracavitary brachytherapy planning for cervical cancer with tandem and ovoid treatments.

Methods

Fifty patients with loco-regionally advanced cervical cancer treated with 200 CT-based tandem and ovoid brachytherapy plans were included. The brachytherapy plans were randomly divided into the training (N = 160) and verification groups (N = 40). The bladder, rectum, sigmoid colon, and small intestine were divided into sub-OARs. The SVM model was established using MATLAB software based on the sub-OAR volume to predict the bladder, rectum, sigmoid colon, and small intestine D 2 c m 3 . Model performance was quantified by mean squared error (MSE) and δ ( δ = | D 2 c m 3 / D prescription ( actual ) - D 2 c m 3 / D prescription ( predicted ) | ) . The goodness of fit of the model was quantified by the coefficient of determination (R²). The accuracy and validity of the SVM model were verified using the validation group.

Results

The D 2 c m 3 value of the bladder, rectum, sigmoid colon, and small intestine correlated with the volume of the corresponding sub-OARs in the training group. The mean squared error (MSE) in the SVM model training group was <0.05; the R² of each OAR was >0.9. There was no significant difference between the D 2 c m 3 -predicted and actual values in the validation group (all P > 0.05): bladder δ = 0.024 ± 0.022, rectum δ = 0.026 ± 0.014, sigmoid colon δ = 0.035 ± 0.023, and small intestine δ = 0.032 ± 0.025.

Conclusion

The SVM model established in this study can effectively predict the D 2 c m 3 for the bladder, rectum, sigmoid colon, and small intestine in cervical cancer brachytherapy.

Integrating Multiomics Information in Deep Learning Architectures for Joint Actuarial Outcome Prediction in Non-Small Cell Lung Cancer Patients After Radiation Therapy

PURPOSE

Novel actuarial deep learning neural network (ADNN) architectures are proposed for joint prediction of radiation therapy outcomes-radiation pneumonitis (RP) and local control (LC)-in stage III non-small cell lung cancer (NSCLC) patients. Unlike normal tissue complication probability/tumor control probability models that use dosimetric information solely, our proposed models consider complex interactions among multiomics information including positron emission tomography (PET) radiomics, cytokines, and miRNAs. Additional time-to-event information is also used in the actuarial prediction.

METHODS AND MATERIALS

Three architectures were investigated: ADNN-DVH considered dosimetric information only; ADNN-com integrated multiomics information; and ADNN-com-joint combined RP2 (RP grade ≥2) and LC prediction. In these architectures, differential dose-volume histograms (DVHs) were fed into 1D convolutional neural networks (CNN) for extracting reduced representations. Variational encoders were used to learn representations of imaging and biological data. Reduced representations were fed into Surv-Nets to predict time-to-event probabilities for RP2 and LC independently and jointly by incorporating time information into designated loss functions.

RESULTS

Models were evaluated on 117 retrospective patients and were independently tested on 25 newly accrued patients prospectively. A multi-institutional RTOG0617 data set of 327 patients was used for external validation. ADNN-DVH yielded cross-validated c-indexes (95% confidence intervals) of 0.660 (0.630-0.690) for RP2 prediction and 0.727 (0.700-0.753) for LC prediction, outperforming a generalized Lyman model for RP2 (0.613 [0.583-0.643]) and a generalized log-logistic model for LC (0.569 [0.545-0.594]). The independent internal test and external validation yielded similar results. ADNN-com achieved an even better performance than ADNN-DVH on both cross-validation and independent internal test. Furthermore, ADNN-com-joint, which yielded performance similar to ADNN-com, realized joint prediction with c-indexes of 0.705 (0.676-0.734) for RP2 and 0.740 (0.714-0.765) for LC and achieved an area under a free-response receiving operator characteristic curve (AU-FROC) of 0.729 (0.697-0.773) for the joint prediction of RP2 and LC.

CONCLUSION

Novel deep learning architectures that integrate multiomics information outperformed traditional normal tissue complication probability/tumor control probability models in actuarial prediction of RP2 and LC.

Estimation of delivered dose to lung tumours considering setup uncertainties and breathing motion in a cohort of patients treated with stereotactic body radiation therapy

INTRODUCTION

Dose-response relationships for local control of lung tumours treated with stereotactic body radiotherapy (SBRT) have proved ambiguous, however, these have been based on the prescribed or planned dose. Delivered dose to the target may be a better predictor for local control. In this study, the probability of the delivered minimum dose to the clinical target volume (CTV) in relation to the prescribed dose was estimated for a cohort of patients, considering geometrical uncertainties.

MATERIALS AND METHODS

Delivered doses were retrospectively simulated for 50 patients treated with SBRT for lung tumours, comparing two image-guidance techniques: pre-treatment verification computed tomography (IG1) and online cone-beam computed tomography (IG2). The prescribed dose was typically to the 67% isodose line of the treatment plan. Simulations used in-house software that shifted the static planned dose according to a breathing motion and sampled setup/matching errors. Each treatment was repeatedly simulated, generating a multiplicity of dose-volume histograms (DVH). From these, tumour-specific and population-averaged statistics were derived.

RESULTS

For IG1, the probability that the minimum CTV dose (D_98%) exceeded 100% of the prescribed dose was 90%. With IG2, this probability increased to 99%.

CONCLUSIONS

Doses below the prescribed dose were delivered to a considerably larger part of the population prior to the introduction of online soft-tissue image-guidance. However, there is no clear evidence that this impacts local control, when compared to previous published data.

Machine learning applications in radiation oncology

Machine learning technology has a growing impact on radiation oncology with an increasing presence in research and industry. The prevalence of diverse data including 3D imaging and the 3D radiation dose delivery presents potential for future automation and scope for treatment improvements for cancer patients. Harnessing this potential requires standardization of tools and data, and focused collaboration between fields of expertise. The rapid advancement of radiation oncology treatment technologies presents opportunities for machine learning integration with investments targeted towards data quality, data extraction, software, and engagement with clinical expertise. In this review, we provide an overview of machine learning concepts before reviewing advances in applying machine learning to radiation oncology and integrating these techniques into the radiation oncology workflows. Several key areas are outlined in the radiation oncology workflow where machine learning has been applied and where it can have a significant impact in terms of efficiency, consistency in treatment and overall treatment outcomes. This review highlights that machine learning has key early applications in radiation oncology due to the repetitive nature of many tasks that also currently have human review. Standardized data management of routinely collected imaging and radiation dose data are also highlighted as enabling engagement in research utilizing machine learning and the ability integrate these technologies into clinical workflow to benefit patients. Physicists need to be part of the conversation to facilitate this technical integration.

Prediction of local recurrence and distant metastasis using radiomics analysis of pretreatment nasopharyngeal [18F]FDG PET/CT images

OBJECTIVES

To develop a radiomics signature to predict locoregional recurrence (LR) and distant metastasis (DM), as extracted from pretreatment 2-deoxy-2-[¹⁸F]fluoro-D-glucose ([18F]FDG) positron emission tomography/X-ray computed tomography (PET/CT) images in locally advanced nasopharyngeal carcinoma (NPC).

MATERIALS AND METHODS

Eighty-five patients with Stage III-IVB NPC underwent pretreatment [¹⁸F]FDG PET/CT scans and received radiotherapy or chemoradiotherapy. 53 of them achieved disease control, and 32 of them failed after treatment (15: LR, 17: DM). A total of 114 radiomic features were extracted from PET/CT images. For univariate analysis, Wilcoxon test and Chi-square test were used to compare median values of features between different treatment outcomes and predict the risk of treatment failure, respectively. For multivariate analysis, all features were grouped into clusters based on Pearson correlation using hierarchical clustering, and the representative feature of each cluster was chosen by the Relief algorithm. Then sequential floating forward selection (SFFS) coupled with a support vector machine (SVM) classifier were used to derive the optimized feature set in terms of the area under receiver operating characteristic (ROC) curve (AUC). The performance of the model was evaluated by leave-one-out-cross-validation, fivefold cross-validation, tenfold cross-validation.

RESULTS

Twenty features had significant differences between disease control and treatment failure. NPC patients with values of Compactness1, Compactness2, Coarseness_NGTDM or SGE_GLGLM above the median as well as patients with values of Irregularity, RLN_GLRLM or GLV_GLSZM below the median, showed a significant (p < 0.05) higher risk of treatment failure (about 50% vs. 25%). The derived radiomics signature consisted of 5 features with the highest AUC value of 0.8290 (sensitivity: 0.8438, specificity: 0.7736) using leave-one-out-cross-validation.

CONCLUSION

Locoregional recurrence (LR) and DM of locally advanced NPC can be predicted using radiomics analysis of pretreatment [¹⁸F]FDG PET/CT. The SFFS feature selection coupled with SVM classifier can derive the optimized feature set with correspondingly highest AUC value for pretreatment prediction of LR and/or DM of NPC.

Patient-specific PTV margins for liver stereotactic body radiation therapy determined using support vector classification with an early warning system for margin adaptation

PURPOSE

An adaptive planning target volume (PTV) margin strategy incorporating a volumetric tracking error assessment after each fraction is proposed for robotic stereotactic body radiation therapy (SBRT) liver treatments.

METHODS AND MATERIALS

A supervised machine learning algorithm employing retrospective data, which emulates a dry-run session prior to planning, is used to investigate if motion tracking errors are <2 mm, and consequently, planning target volume (PTV) margins can be reduced. A fraction of data collected during the beginning of a treatment course emulates a dry-run session (mock) before planning. Twenty features are calculated using mock data and used for support vector classification (SVC). A treatment course is labeled as Class 1 if the maximum root-mean-square radial tracking error for all remaining fractions is below 2 mm, or Class 2 otherwise. We evaluate the classification using fivefold cross-validation, leave-one-out cross-validation, 500 repeated random subsampling cross-validation, and the receiver operating characteristic (ROC) metric. The classification is independently cross-validated on a cohort of 48 treatment plans for other anatomical sites. A per fraction assessment of volumetric tracking errors is performed for the standard 5 mm PTV margin (PTV_std ) for courses predicted as Class 2; or for a margin reduced by 2 mm (PTV_std-2mm ) for those predicted as Class 1. We perturb the gross tumor volume (GTV) by the tracking errors for each x-ray image acquisition and calculate the fractional GTV voxel occupancy probability (P_i ) inside the PTV for each treatment fraction i. For treatment courses classified as Class 1, an early warning system flags treatment courses having any P_i  < 0.99, and the subsequent treatments are proposed to be replanned using PTV_std .

RESULTS

The classification accuracies are 0.84 ± 0.06 using fivefold cross-validation, and 0.77 when validated using an independent testing set (other anatomical sites). Eighty percent of treatment courses are correctly classified using leave-one-out cross-validation. The sensitivity, precision, specificity, F1 score, and accuracy are 0.81 ± 0.09, 0.85 ± 0.08, 0.80 ± 0.11, 0.83 ± 0.06, and 0.80 ± 0.07, respectively, using 500 repeated random subsampling cross-validation. The area under the curve for the ROC metric is 0.87 ± 0.05. The four most important features for classification are related to standard deviations of motion tracking errors, the linearity between the target location and external LED marker positions, and marker radial motion amplitudes. Eleven of 64 cases predicted to be of Class 1 have 0.96 < P_i  < 0.99 for each treatment fraction, and require replanning using PTV_std . In comparison, the PTV_std always covers the perturbed GTVs with P_i  > 0.99 for all patients.

CONCLUSIONS

Support vector classification is proposed for the classification of different motion tracking errors for patient courses based on a mock session before planning for SBRT liver treatments. It is feasible to implement patient-specific PTV margins in the clinic, assisted with an early warning system to flag treatment courses that require replanning using larger PTV margins in an adaptive treatment strategy.

Artificial Intelligence in Lung Cancer: Bridging the Gap Between Computational Power and Clinical Decision-Making

Lung cancer remains the most common cause of cancer death worldwide. Recent advances in lung cancer screening, radiotherapy, surgical techniques, and systemic therapy have led to increasing complexity in diagnosis, treatment decision-making, and assessment of recurrence. Artificial intelligence (AI)-based prediction models are being developed to address these issues and may have a future role in screening, diagnosis, treatment selection, and decision-making around salvage therapy. Imaging plays an essential role in all components of lung cancer management and has the potential to play a key role in AI applications. Artificial intelligence has demonstrated value in prognostic biomarker discovery in lung cancer diagnosis, treatment, and response assessment, putting it at the forefront of the next phase of personalized medicine. However, although exploratory studies demonstrate potential utility, there is a need for rigorous validation and standardization before AI can be utilized in clinical decision-making. In this review, we will provide a summary of the current literature implementing AI for outcome prediction in lung cancer. We will describe the anticipated impact of AI on the management of patients with lung cancer and discuss the challenges of clinical implementation of these techniques.

Estimation of the α/β ratio of non-small cell lung cancer treated with stereotactic body radiotherapy

BACKGROUND

High-dose hypofractionated radiotherapy should theoretically result in a deviation from the typical linear-quadratic shape of the cell survival curve beyond a certain threshold dose, yet no evidence for this hypothesis has so far been found in clinical data of stereotactic body radiotherapy treatment (SBRT) for early-stage non-small cell lung cancer (NSCLC). A pragmatic explanation is a larger α/β ratio than the conventionally assumed 10 Gy. We here attempted an estimation of the α/β ratio for NSCLC treated with SBRT using individual patient data.

MATERIALS AND METHODS

We combined two large retrospective datasets, yielding 1294 SBRTs (≤10 fractions) of early stage NSCLC. Cox proportional hazards regression, a logistic tumor control probability model and a biologically motivated Bayesian cure rate model were used to estimate the α/β ratio based on the observed number of local recurrences and accounting for tumor size.

RESULTS

A total of 109 local progressions were observed after a median of 17.7 months (range 0.6-76.3 months). Cox regression, logistic regression of 3 year tumor control probability and the cure rate model yielded best-fit estimates of α/β = 12.8 Gy, 14.9 Gy and 12-16 Gy (depending on the prior for α/β), respectively, although with large uncertainties that did not rule out the conventional α/β = 10 Gy.

CONCLUSIONS

Clinicians can continue to use the simple LQ formalism to compare different SBRT treatment schedules for NSCLC. While α/β = 10 Gy is not ruled out by our data, larger values in the range 12-16 Gy are more probable, consistent with recent meta-regression analyses.

Balancing accuracy and interpretability of machine learning approaches for radiation treatment outcomes modeling

Radiation outcomes prediction (ROP) plays an important role in personalized prescription and adaptive radiotherapy. A clinical decision may not only depend on an accurate radiation outcomes’ prediction, but also needs to be made based on an informed understanding of the relationship among patients’ characteristics, radiation response and treatment plans. As more patients’ biophysical information become available, machine learning (ML) techniques will have a great potential for improving ROP. Creating explainable ML methods is an ultimate task for clinical practice but remains a challenging one. Towards complete explainability, the interpretability of ML approaches needs to be first explored. Hence, this review focuses on the application of ML techniques for clinical adoption in radiation oncology by balancing accuracy with interpretability of the predictive model of interest. An ML algorithm can be generally classified into an interpretable (IP) or non-interpretable (NIP) (“black box”) technique. While the former may provide a clearer explanation to aid clinical decision-making, its prediction performance is generally outperformed by the latter. Therefore, great efforts and resources have been dedicated towards balancing the accuracy and the interpretability of ML approaches in ROP, but more still needs to be done. In this review, current progress to increase the accuracy for IP ML approaches is introduced, and major trends to improve the interpretability and alleviate the “black box” stigma of ML in radiation outcomes modeling are summarized. Efforts to integrate IP and NIP ML approaches to produce predictive models with higher accuracy and interpretability for ROP are also discussed.

ICRU report 91 on prescribing, recording, and reporting of stereotactic treatments with small photon beams : Statement from the DEGRO/DGMP working group stereotactic radiotherapy and radiosurgery

The International Commission on Radiation Units and Measurements (ICRU) report 91 with the title "prescribing, recording, and reporting of stereotactic treatments with small photon beams" was published in 2017. This extensive publication covers different relevant aspects of stereotactic radiotherapy such as small field dosimetry, accuracy requirements for volume definition and planning algorithms, and the precise application of treatment by means of image guidance. Finally, recommendations for prescribing, recording and reporting are given.

Assessing the Role of Artificial Intelligence (AI) in Clinical Oncology: Utility of Machine Learning in Radiotherapy Target Volume Delineation

The fields of radiotherapy and clinical oncology have been rapidly changed by the advances of technology. Improvement in computer processing power and imaging quality heralded precision radiotherapy allowing radiotherapy to be delivered efficiently, safely and effectively for patient benefit. Artificial intelligence (AI) is an emerging field of computer science which uses computer models and algorithms to replicate human-like intelligence and perform specific tasks which offers a huge potential to healthcare. We reviewed and presented the history, evolution and advancement in the fields of radiotherapy, clinical oncology and machine learning. Radiotherapy target delineation is a complex task of outlining tumour and organ at risks volumes to allow accurate delivery of radiotherapy. We discussed the radiotherapy planning, treatment delivery and reviewed how technology can help with this challenging process. We explored the evidence and clinical application of machine learning to radiotherapy. We concluded on the challenges, possible future directions and potential collaborations to achieve better outcome for cancer patients.

Gantry angle classification with a fluence map in intensity-modulated radiotherapy for prostate cases using machine learning

We investigated the gantry-angle classifier performance with a fluence map using three machine-learning algorithms, and compared it with human performance. Eighty prostate cases were investigated using a seven-field-intensity modulated radiotherapy treatment (IMRT) plan with beam angles of 0°, 50°, 100°, 155°, 205°, 260°, and 310°. The k-nearest neighbor (k-NN), logistic regression (LR), and support vector machine (SVM) algorithms were used. In the observer test, three radiotherapists assessed the gantry angle classification in a blind manner. The precision and recall rates were calculated for the machine learning and observer test. The average precision rate of the k-NN and LR algorithms were 94.8% and 97.9%, respectively. The average recall rate of the k-NN and LR algorithms were 94.3% and 97.9%, respectively. The SVM had 100% precision and recall rates. The gantry angles of 0°, 155°, and 205° had an accuracy of 100% in all algorithms. In the observer test, average precision and recall rates were 82.6% and 82.6%, respectively. All observers could easily classify the gantry angles of 0°, 155°, and 205° with a high degree of accuracy. Misclassifications occurred in gantry angles of 50°, 100°, 260°, and 310°. Machine learning could better classify gantry angles for prostate IMRT than human beings. In particular, the SVM algorithm had a perfect classification of 100%.

Data Analysis in Radiotherapy Treatments

Radiotherapy is one of the main cancer treatments available today, together with chemotherapy and surgery. Radiotherapy treatments have to be planned for each patient in an individualized manner. The knowledge acquired from one single treatment can be used to improve the treatment planning and outcome of several other patients. In the last years, attention has been drawn to the added value of using data analysis for radiotherapy treatment planning, prediction of treatment outcomes, survival analysis and quality assurance. In this article, existing literature is reviewed.

Clinical Results of Mean GTV Dose Optimized Robotic-Guided Stereotactic Body Radiation Therapy for Lung Tumors

Introduction

We retrospectively evaluated the efficacy and toxicity of gross tumor volume (GTV) mean dose optimized stereotactic body radiation therapy (SBRT) for primary and secondary lung tumors with and without robotic real-time motion compensation.

Materials and methods

Between 2011 and 2017, 208 patients were treated with SBRT for 111 primary lung tumors and 163 lung metastases with a median GTV of 8.2 cc (0.3–174.0 cc). Monte Carlo dose optimization was performed prioritizing GTV mean dose at the potential cost of planning target volume (PTV) coverage reduction while adhering to safe normal tissue constraints. The median GTV mean biological effective dose (BED)₁₀ was 162.0 Gy₁₀ (34.2–253.6 Gy₁₀) and the prescribed PTV BED₁₀ ranged 23.6–151.2 Gy₁₀ (median, 100.8 Gy₁₀). Motion compensation was realized through direct tracking (44.9%), fiducial tracking (4.4%), and internal target volume (ITV) concepts with small (≤5 mm, 33.2%) or large (>5 mm, 17.5%) motion. The local control (LC), progression-free survival (PFS), overall survival (OS), and toxicity were analyzed.

Results

Median follow-up was 14.5 months (1–72 months). The 2-year actuarial LC, PFS, and OS rates were 93.1, 43.2, and 62.4%, and the median PFS and OS were 18.0 and 39.8 months, respectively. In univariate analysis, prior local irradiation (hazard ratio (HR) 0.18, confidence interval (CI) 0.05–0.63, p = 0.01), GTV/PTV (HR 1.01–1.02, CI 1.01–1.04, p < 0.02), and PTV prescription, mean GTV, and maximum plan BED₁₀ (HR 0.97–0.99, CI 0.96–0.99, p < 0.01) were predictive for LC while the tracking method was not (p = 0.97). For PFS and OS, multivariate analysis showed Karnofsky Index (p < 0.01) and tumor stage (p ≤ 0.02) to be significant factors for outcome prediction. Late radiation pneumonitis or chronic rip fractures grade 1–2 were observed in 5.3% of the patients. Grade ≥3 side effects did not occur.

Conclusion

Robotic SBRT is a safe and effective treatment for lung tumors. Reducing the PTV prescription and keeping high GTV mean doses allowed the reduction of toxicity while maintaining high local tumor control. The use of real-time motion compensation is strongly advised, however, well-performed ITV motion compensation may be used alternatively when direct tracking is not feasible.

Radiobiological Optimization in Lung Stereotactic Body Radiation Therapy: Are We Ready to Apply Radiobiological Models?

Lung tumors are often associated with a poor prognosis although different schedules and treatment modalities have been extensively tested in the clinical practice. The complexity of this disease and the use of combined therapeutic approaches have been investigated and the use of high dose-rates is emerging as effective strategy. Technological improvements of clinical linear accelerators allow combining high dose-rate and a more conformal dose delivery with accurate imaging modalities pre- and during therapy. This paper aims at reporting the state of the art and future direction in the use of radiobiological models and radiobiological-based optimizations in the clinical practice for the treatment of lung cancer. To address this issue, a search was carried out on PubMed database to identify potential papers reporting tumor control probability and normal tissue complication probability for lung tumors. Full articles were retrieved when the abstract was considered relevant, and only papers published in English language were considered. The bibliographies of retrieved papers were also searched and relevant articles included. At the state of the art, dose–response relationships have been reported in literature for local tumor control and survival in stage III non-small cell lung cancer. Due to the lack of published radiobiological models for SBRT, several authors used dose constraints and models derived for conventional fractionation schemes. Recently, several radiobiological models and parameters for SBRT have been published and could be used in prospective trials although external validations are recommended to improve the robustness of model predictive capability. Moreover, radiobiological-based functions have been used within treatment planning systems for plan optimization but the advantages of using this strategy in the clinical practice are still under discussion. Future research should be directed toward combined regimens, in order to potentially improve both local tumor control and survival. Indeed, accurate knowledge of the relevant parameters describing tumor biology and normal tissue response is mandatory to correctly address this issue. In this context, the role of medical physicists and the AAPM in the development of radiobiological models is crucial for the progress of developing specific tool for radiobiological-based optimization treatment planning.

[Rectal toxicity prediction based on accurate rectal surface dose summation for cervical cancer radiotherapy]

OBJECTIVE

To propose arectal toxicity prediction method based on deformable surface dose accumulation.

METHODS

The clinical data were collected retrospectively from 42patients receiving radiotherapy for cervical cancer. With the first fraction as the reference, the other fractions of rectum surface were registered to the reference fraction to obtain the deformation vector fields (DVFs), which were used to deform and sum the fractional rectal doses to yield the cumulative rectal dose. The cumulative rectal dose was flattened via 3D-2D mapping to generate a 2D rectum surface dose map. Two dosimetric features, namely DVPs and DGPs were extracted. Logistic regression embedded with sequential forward feature selection was used as the prediction model. The predictive performance was evaluated in terms of the accuracy, sensitivity, specificity, and the area under the receiver operating characteristic (ROC) curve (AUC).

RESULTS

Significant improvements for rectum surface DIR were achieved. The best predictive results were achieved by using both DVPs and DGPs as the features with a sensitivity of 79.5%, a specificity of 81.3% and an AUC of 0.88.

CONCLUSION

The proposed method is feasible for predicting clinical rectal toxicity in patients undergoing radiotherapy for cervical cancer.

Biological function integrated prediction of severe radiographic progression in rheumatoid arthritis: a nested case control study

Background

Radiographic progression is reported to be highly heritable in rheumatoid arthritis (RA). However, previous study using genetic loci showed an insufficient accuracy of prediction for radiographic progression. The aim of this study is to identify a biologically relevant prediction model of radiographic progression in patients with RA using a genome-wide association study (GWAS) combined with bioinformatics analysis.

Methods

We obtained genome-wide single nucleotide polymorphism (SNP) data for 374 Korean patients with RA using Illumina HumanOmni2.5Exome-8 arrays. Radiographic progression was measured using the yearly Sharp/van der Heijde modified score rate, and categorized in no or severe progression. Significant SNPs for severe radiographic progression from GWAS were mapped on the functional genes and reprioritized by post-GWAS analysis. For robust prediction of radiographic progression, tenfold cross-validation using a support vector machine (SVM) classifier was conducted. Accuracy was used for selection of optimal SNPs set in the Hanyang Bae RA cohort. The performance of our final model was compared with that of other models based on GWAS results and SPOT (one of the post-GWAS analyses) using receiver operating characteristic (ROC) curves. The reliability of our model was confirmed using GWAS data of Caucasian patients with RA.

Results

A total of 36,091 significant SNPs with a p value <0.05 from GWAS were reprioritized using post-GWAS analysis and approximately 2700 were identified as SNPs related to RA biological features. The best average accuracy of ten groups was 0.6015 with 85 SNPs, and this increased to 0.7481 when combined with clinical information. In comparisons of the performance of the model, the 0.7872 area under the curve (AUC) in our model was superior to that obtained with GWAS (AUC 0.6586, p value 8.97 × 10^-5) or SPOT (AUC 0.7449, p value 0.0423). Our model strategy also showed superior prediction accuracy in Caucasian patients with RA compared with GWAS (p value 0.0049) and SPOT (p value 0.0151).

Conclusions

Using various biological functions of SNPs and repeated machine learning, our model could predict severe radiographic progression relevantly and robustly in patients with RA compared with models using only GWAS results or other post-GWAS tools.

Electronic supplementary material

The online version of this article (doi:10.1186/s13075-017-1414-x) contains supplementary material, which is available to authorized users.

Breathing-motion-compensated robotic guided stereotactic body radiation therapy : Patterns of failure analysis

PURPOSE

We retrospectively evaluated the patterns of failure for robotic guided real-time breathing-motion-compensated (BMC) stereotactic body radiation therapy (SBRT) in the treatment of tumors in moving organs.

PATIENTS AND METHODS

Between 2011 and 2016, a total of 198 patients with 280 lung, liver, and abdominal tumors were treated with BMC-SBRT. The median gross tumor volume (GTV) was 12.3 cc (0.1-372.0 cc). Medians of mean GTV BED_{α/β =10 Gy} (BED = biological effective dose) was 148.5 Gy₁₀ (31.5-233.3 Gy₁₀) and prescribed planning target volume (PTV) BED_{α/β =10 Gy} was 89.7 Gy₁₀ (28.8-151.2 Gy₁₀), respectively. We analyzed overall survival (OS) and local control (LC) based on various factors, including BEDs with α/β ratios of 15 Gy (lung metastases), 21 Gy (primary lung tumors), and 27 Gy (liver metastases).

RESULTS

Median follow-up was 10.4 months (2.0-59.0 months). The 2‑year actuarial LC was 100 and 86.4% for primary early and advanced stage lung tumors, respectively, 100% for lung metastases, 82.2% for liver metastases, and 90% for extrapulmonary extrahepatic metastases. The 2‑year OS rate was 47.9% for all patients. In uni- and multivariate analysis, comparatively lower PTV prescription dose (equivalence of 3 × 12-13 Gy) and higher average GTV dose (equivalence of 3 × 18 Gy) to current practice were significantly associated with LC. For OS, Karnofsky performance score (100%), gender (female), and SBRT without simultaneous chemotherapy were significant prognostic factors. Grade 3 side effects were rare (0.5%).

CONCLUSIONS

Robotic guided BMC-SBRT can be considered a safe and effective treatment for solid tumors in moving organs. To reach sufficient local control rates, high average GTV doses are necessary. Further prospective studies are warranted to evaluate these points.

Stereotactic body radiotherapy for oligo-metastatic liver disease - Influence of pre-treatment chemotherapy and histology on local tumor control

INTRODUCTION

Stereotactic body radiation therapy (SBRT) is applied in the oligometastatic setting to treat liver metastases. However, factors influencing tumor control probability (TCP) other than radiation dose have not been thoroughly investigated. Here we set out to investigate such factors with a focus on the influence of histology and chemotherapy prior to SBRT using a large multi-center database from the German Society of Radiation Oncology.

METHODS

452 SBRT treatments in 363 patients were analyzed after collection of patient, tumor and treatment data in a multi-center database. Histology was considered through random effects in semi-parametric and parametric frailty models. Dose prescriptions were parametrized by conversion to the maximum biologically effective dose using alpha/beta of 10Gy (BED_max).

RESULTS

After adjusting for histology, BED_max was the strongest predictor of TCP. Larger PTV volumes, chemotherapy prior to SBRT and simple motion management techniques predicted significantly lower TCP. The model predicted a BED of 209±67Gy₁₀ necessary for 90% TCP at 2years with no prior chemotherapy, but 286±78Gy₁₀ when chemotherapy had been given. Breast cancer metastases were significantly more responsive to SBRT compared to other histologies with 90% TCP at 2years achievable with BED_max of 157±80Gy₁₀ or 80±62Gy₁₀ with and without prior chemotherapy, respectively.

CONCLUSIONS

Besides dose, histology and pretreatment chemotherapy were important factors influencing local TCP in this large cohort of liver metastases. After adjusting for prior chemotherapy, our data add to the emerging evidence that breast cancer metastases do respond better to hypofractionated SBRT compared to other histologies.

Using machine learning to predict radiation pneumonitis in patients with stage I non-small cell lung cancer treated with stereotactic body radiation therapy

To develop a patient-specific 'big data' clinical decision tool to predict pneumonitis in stage I non-small cell lung cancer (NSCLC) patients after stereotactic body radiation therapy (SBRT). 61 features were recorded for 201 consecutive patients with stage I NSCLC treated with SBRT, in whom 8 (4.0%) developed radiation pneumonitis. Pneumonitis thresholds were found for each feature individually using decision stumps. The performance of three different algorithms (Decision Trees, Random Forests, RUSBoost) was evaluated. Learning curves were developed and the training error analyzed and compared to the testing error in order to evaluate the factors needed to obtain a cross-validated error smaller than 0.1. These included the addition of new features, increasing the complexity of the algorithm and enlarging the sample size and number of events. In the univariate analysis, the most important feature selected was the diffusion capacity of the lung for carbon monoxide (DLCO adj%). On multivariate analysis, the three most important features selected were the dose to 15 cc of the heart, dose to 4 cc of the trachea or bronchus, and race. Higher accuracy could be achieved if the RUSBoost algorithm was used with regularization. To predict radiation pneumonitis within an error smaller than 10%, we estimate that a sample size of 800 patients is required. Clinically relevant thresholds that put patients at risk of developing radiation pneumonitis were determined in a cohort of 201 stage I NSCLC patients treated with SBRT. The consistency of these thresholds can provide radiation oncologists with an estimate of their reliability and may inform treatment planning and patient counseling. The accuracy of the classification is limited by the number of patients in the study and not by the features gathered or the complexity of the algorithm.

Big Data and machine learning in radiation oncology: State of the art and future prospects

Precision medicine relies on an increasing amount of heterogeneous data. Advances in radiation oncology, through the use of CT Scan, dosimetry and imaging performed before each fraction, have generated a considerable flow of data that needs to be integrated. In the same time, Electronic Health Records now provide phenotypic profiles of large cohorts of patients that could be correlated to this information. In this review, we describe methods that could be used to create integrative predictive models in radiation oncology. Potential uses of machine learning methods such as support vector machine, artificial neural networks, and deep learning are also discussed.

Symbolic features and classification via support vector machine for predicting death in patients with Chagas disease

This paper introduces a technique for predicting death in patients with Chagas disease using features extracted from symbolic series and time-frequency indices of heart rate variability (HRV). The study included 150 patients: 15 patients who died and 135 who did not. The HRV series were obtained from 24-h Holter monitoring. Sequences of symbols from 5-min epochs from series of RR intervals were generated using symbolic dynamics and ordinal pattern statistics. Fourteen features were extracted from symbolic series and four derived from clinical aspects of patients. For classification, the 18 features from each epoch were used as inputs in a support vector machine (SVM) with a radial basis function (RBF) kernel. The results showed that it is possible to distinguish between the two classes, patients with Chagas disease who did or did not die, with a 95% accuracy rate. Therefore, we suggest that the use of new features based on symbolic series, coupled with classic time-frequency and clinical indices, proves to be a good predictor of death in patients with Chagas disease.

Fusion of clinical and stochastic finite element data for hip fracture risk prediction

Hip fracture affects more than 250,000 people in the US and 1.6 million worldwide per year. With an aging population, the development of reliable fracture risk models is therefore of prime importance. Due to the complexity of the hip fracture phenomenon, the use of clinical data only, as it is done traditionally, might not be sufficient to ensure an accurate and robust hip fracture prediction model. In order to increase the predictive ability of the risk model, the authors propose to supplement the clinical data with computational data from finite element models. The fusion of the two types of data is performed using deterministic and stochastic computational data. In the latter case, uncertainties in loading and material properties of the femur are accounted for and propagated through the finite element model. The predictive capability of a support vector machine (SVM) risk model constructed by combining clinical and finite element data was assessed using a Women׳s Health Initiative (WHI) dataset. The dataset includes common factors such as age and BMD as well as geometric factors obtained from DXA imaging. The fusion of computational and clinical data systematically leads to an increase in predictive ability of the SVM risk model as measured by the AUC metric. It is concluded that the largest gains in AUC are obtained by the stochastic approach. This gain decreases as the dimensionality of the problem increases: a 5.3% AUC improvement was achieved for a 9 dimensional problem involving geometric factors and weight while a 1.3% increase was obtained for a 20 dimensional case including geometric and conventional factors.

Machine Learning Approaches for Predicting Radiation Therapy Outcomes: A Clinician's Perspective

Radiation oncology has always been deeply rooted in modeling, from the early days of isoeffect curves to the contemporary Quantitative Analysis of Normal Tissue Effects in the Clinic (QUANTEC) initiative. In recent years, medical modeling for both prognostic and therapeutic purposes has exploded thanks to increasing availability of electronic data and genomics. One promising direction that medical modeling is moving toward is adopting the same machine learning methods used by companies such as Google and Facebook to combat disease. Broadly defined, machine learning is a branch of computer science that deals with making predictions from complex data through statistical models. These methods serve to uncover patterns in data and are actively used in areas such as speech recognition, handwriting recognition, face recognition, "spam" filtering (junk email), and targeted advertising. Although multiple radiation oncology research groups have shown the value of applied machine learning (ML), clinical adoption has been slow due to the high barrier to understanding these complex models by clinicians. Here, we present a review of the use of ML to predict radiation therapy outcomes from the clinician's point of view with the hope that it lowers the "barrier to entry" for those without formal training in ML. We begin by describing 7 principles that one should consider when evaluating (or creating) an ML model in radiation oncology. We next introduce 3 popular ML methods--logistic regression (LR), support vector machine (SVM), and artificial neural network (ANN)--and critique 3 seminal papers in the context of these principles. Although current studies are in exploratory stages, the overall methodology has progressively matured, and the field is ready for larger-scale further investigation.

Safety of high-dose-rate stereotactic body radiotherapy

Background and purpose

Flattening filter free (FFF) beams with high dose rate are increasingly used for stereotactic body radiotherapy (SBRT), because they substantially shorten beam-on time. The physical properties of these beams together with potentially unknown radiobiological effects might affect patient safety. Therefore here we analyzed the clinical outcome of our patients.

Material and methods

Between 3/2010 and 2/2014 84 patients with 100 lesions (lung 75, liver 10, adrenal 6, lymph nodes 5, others 4) were treated with SBRT using 6 MV FFF or 10 MV FFF beams at our institution. Clinical efficacy endpoints and toxicity were assessed by Kaplan-Meier analysis and CTCAE criteria version 4.0.

Results

Median follow-up was 11 months (range: 3–41). No severe acute toxicity was observed. There has been one case of severe late toxicity (1%), a grade 3 bile duct stricture that was possibly related to SBRT. For all patients, the 1-year local control rate, progression free survival and overall survival were 94%, 38% and 80% respectively, and for patients with lung lesions 94%, 48% and 83%, respectively.

Conclusions

No unexpected toxicity occurred. Toxicity and treatment efficacy are perfectly in range with studies investigating SBRT with flattened beams. The use of FFF beams at maximum dose rate for SBRT is time efficient and appears to be safe.

Three-dimensional SVM with latent variable: application for detection of lung lesions in CT images

The study aims to improve the performance of current computer-aided schemes for the detection of lung lesions, especially the low-contrast in gray density or irregular in shape. The relative position between suspected lesion and whole lung is, for the first time, added as a latent feature to enrich current Three-dimensional (3D) features such as shape, texture. Subsequently, 3D matrix patterns-based Support Vector Machine (SVM) with the latent variable, referred to as L-SVM3Dmatrix, was constructed accordingly. A CT image database containing 750 abnormal cases with 1050 lesions was used to train and evaluate several similar computer-aided detection (CAD) schemes: traditional features-based SVM (SVMfeature), 3D matrix patterns-based SVM (SVM3Dmatrix) and L-SVM3Dmatrix. The classifier performances were evaluated by computing the area under the ROC curve (AUC), using a 5-fold cross-validation. The L-SVM3Dmatrix sensitivity was 93.0 with 1.23% percentage of False Positive (FP), the SVM3Dmatrix sensitivity was 88.4 with 1.49% percentage of FP, and the SVMfeature sensitivity was 87.2 with 1.78% percentage of FP. The L-SVM3Dmatrix outperformed other current lung CAD schemes, especially regarding the difficult lesions.

Machine learning methods in the computational biology of cancer

The objectives of this Perspective paper are to review some recent advances in sparse feature selection for regression and classification, as well as compressed sensing, and to discuss how these might be used to develop tools to advance personalized cancer therapy. As an illustration of the possibilities, a new algorithm for sparse regression is presented and is applied to predict the time to tumour recurrence in ovarian cancer. A new algorithm for sparse feature selection in classification problems is presented, and its validation in endometrial cancer is briefly discussed. Some open problems are also presented.