Brain metastases are a frequent and debilitating manifestation of advanced cancer. Here, we collect and analyze neuroimaging of 3,065 cancer patients with 13,067 brain metastases, representing an extensive collection for research. We find that metastases predominantly localize to high perfusion areas near the grey-white matter junction, but also identify notable differences depending on the primary cancer histology as well as brain regions which do not conform to this relationship. Lung and breast cancers, in contrast to melanoma, frequently metastasize to the cerebellum, hinting at biological pathways of spread. Additionally, the deep brain structures are relatively spared from metastasis, regardless of primary cancer type. Leveraging this data, we propose a probabilistic brain metastasis risk model to enhance the therapeutic ratio of whole-brain radiotherapy by targeting high risk areas while preserving cortical and subcortical brain regions of functional significance and low metastasis risk, potentially reducing the cognitive side effects of therapy.

Background Evaluation of treatment response for brain metastases (BMs) following stereotactic radiosurgery (SRS) becomes complex as the number of treated BMs increases. This study uses artificial intelligence (AI) to track BMs after SRS and validates its output compared with manual measurements.Methods Patients with BMs who received at least one course of SRS and followed up with MRI scans were retrospectively identified. A tool for automated detection, segmentation, and tracking of intracranial metastases on longitudinal imaging, MEtastasis Tracking with Repeated Observations (METRO), was applied to the dataset. The longest three-dimensional (3D) diameter identified with METRO was compared with manual measurements of maximum axial BM diameter, and their correlation was analyzed. Change in size of the measured BM identified with METRO after SRS treatment was used to classify BMs as responding, or not responding, to treatment, and its accuracy was determined relative to manual measurements.Results From 71 patients, 176 BMs were identified and measured with METRO and manual methods. Based on a one-to-one correlation analysis, the correlation coefficient was R2 = 0.76 (P = .0001). Using modified BM response classifications of BM change in size, the longest 3D diameter data identified with METRO had a sensitivity of 0.72 and a specificity of 0.95 in identifying lesions that responded to SRS, when using manual axial diameter measurements as the ground truth.Conclusions Using AI to automatically measure and track BM volumes following SRS treatment, this study showed a strong correlation between AI-driven measurements and the current clinically used method: manual axial diameter measurements.

Background and purpose: Patients with brain metastases (BMs) are surviving longer and returning for multiple courses of stereotactic radiosurgery. BMs are monitored after radiation with follow-up magnetic resonance (MR) imaging every 2-3 months. This study investigated whether it is possible to automatically track BMs on longitudinal imaging and quantify the tumor response after radiotherapy. Methods: The METRO process (MEtastasis Tracking with Repeated Observations was developed to automatically process patient data and track BMs. A longitudinal intrapatient registration method for T1 MR post-Gd was conceived and validated on 20 patients. Detections and volumetric measurements of BMs were obtained from a deep learning model. BM tracking was validated on 32 separate patients by comparing results with manual measurements of BM response and radiologists' assessments of new BMs. Linear regression and residual analysis were used to assess accuracy in determining tumor response and size change. Results: A total of 123 irradiated BMs and 38 new BMs were successfully tracked. 66 irradiated BMs were visible on follow-up imaging 3-9 months after radiotherapy. Comparing their longest diameter changes measured manually vs. METRO, the Pearson correlation coefficient was 0.88 (p < 0.001); the mean residual error was 8 +/- 17%. The mean registration error was 1.5 +/- 0.2 mm. Conclusions: Automatic, longitudinal tracking of BMs using deep learning methods is feasible. In particular, the software system METRO fulfills a need to automatically track and quantify volumetric changes of BMs prior to, and in response to, radiation therapy.

Purpose/Objective(s): Modern therapeutics for patients with metastatic solid tumors have resulted in improved systemic disease control and overall survival, but these drugs are not generally CNS-active.Thus, the control of brain metastasis (BM) with local therapies such as stereotactic radiosurgery (SRS) has become increasingly important.The current gold standard for evaluating BM response is manual measurement of the maximum axial diameter, a process that becomes more complex as the number of BMs increases.Artificial intelligence (AI) can be used to automate this laborious process and open big data analytics.In this study, BM measurements obtained by AI were compared to the manual human-collected data.Materials/Methods: From an institutional database, patients with renal-cell carcinoma (RCC) BM who received SRS were retrospectively identified.A tool developed in-house for automated detection and segmentation of intracranial metastases on longitudinal images, the BMTracker, was applied to the dataset.The BMTracker utilizes a 3D convolutional neural network to segment gross tumor volumes (GTVs) on pre-and post-treatment follow-up T1 MR images and automated rigid registration to identify and track each lesion on longitudinal scans.The software then calculates the BM volume and Feret diameter over all available scans following SRS.The Feret diameter is defined as the diameter of the sphere with the minimum volume needed to encompass the lesion in three-dimensions.For evaluation, the correlation between the manual and automated lesion measurements was obtained and the two measurements were compared at each time point.Results: From 65 patients with RCC BMs, 164 lesions were identified and measured with the BMTracker and manual methods.The pre-SRS treatment MR was median 14.0 days [-18.0,-9.8] before SRS treatment.The AIderived Feret diameter on the pre-treatment scan was median 11.7 mm [8.0, 19.2], while manual axial diameter measurement was 7.5 mm [4.9, 13.2].There were n = 100 BMs with at least one follow-up MR (median = 55 days after SRS [40, 84]) and n = 71 with at least two followup MRs (median = 145 days after SRS [96, 200]).At first and second follow-up after SRS treatment, AI-derived Feret diameter was median 9.6 mm [4.6, 18.5] and 10.2 mm [0, 19.8], respectively, while manual axial diameter was median 6.3 mm [0.9, 12] and 6.0 mm [0, 14.9], respectively.Based on a one-to-one correlation analysis between the lesion Feret diameter determined by the BMTracker, and the manually measured lesion axial diameter, the correlation coefficient was R 2 = 0.789 (p = 0.0001).Conclusion: Using AI to automatically track and measure BM volumes following SRS treatment, this study found a strong correlation between lesion diameter based on AI-driven measurements and the current gold standard.AI allows efficient and accurate segmentation and tracking of BMs.Future studies will compare the BMTracker data to clinical response data.

An increasing number of patients with multiple brain metastases are being treated with stereotactic radiosurgery (SRS). Manually identifying and contouring all metastatic lesions is difficult and time-consuming, and a potential source of variability. Hence, we developed a 3D deep learning approach for segmenting brain metastases on MR and CT images. Five-hundred eleven patients treated with SRS were retrospectively identified for this study. Prior to radiotherapy, the patients were imaged with 3D T1 spoiled-gradient MR post-Gd (T1 + C) and contrast-enhanced CT (CECT), which were co-registered by a treatment planner. The gross tumor volume contours, authored by the attending radiation oncologist, were taken as the ground truth. There were 3 +/- 4 metastases per patient, with volume up to 57 ml. We produced a multi-stage model that automatically performs brain extraction, followed by detection and segmentation of brain metastases using co-registered T1 + C and CECT. Augmented data from 80% of these patients were used to train modified 3D V-Net convolutional neural networks for this task. We combined a normalized boundary loss function with soft Dice loss to improve the model optimization, and employed gradient accumulation to stabilize the training. The average Dice similarity coefficient (DSC) for brain extraction was 0.975 +/- 0.002 (95% CI). The detection sensitivity per metastasis was 90% (329/367), with moderate dependence on metastasis size. Averaged across 102 test patients, our approach had metastasis detection sensitivity 95 +/- 3%, 2.4 +/- 0.5 false positives, DSC of 0.76 +/- 0.03, and 95th-percentile Hausdorff distance of 2.5 +/- 0.3 mm (95% CIs). The volumes of automatic and manual segmentations were strongly correlated for metastases of volume up to 20 ml ( r=0.97, p<0.001 <i ). This work expounds a fully 3D deep learning approach capable of automatically detecting and segmenting brain metastases using co-registered T1 + C and CECT.