Toward automatic beam angle selection for pencil-beam scanning proton liver Treatments: A deep learning-based approach

BACKGROUND

Dose deposition characteristics of proton radiation can be advantageous over photons. Proton treatment planning however poses additional challenges for the planners. Proton therapy is usually delivered with only a small number of beam angles, and the quality of a proton treatment plan is largely determined by the beam angles employed. Finding the optimal beam angles for a proton treatment plan requires time and experience, motivating the investigation of automatic beam angle selection methods.

PURPOSE

A deep learning-based approach to automatic beam angle selection is proposed for proton pencil-beam scanning treatment planning of liver lesions.

METHODS

We cast beam-angle selection as a multi-label classification problem. To account for angular boundary discontinuity, the underlying convolution neural network is trained with the proposed Circular Earth Mover's Distance based regularization and multi-label circular-smooth label technique. Furthermore, an analytical algorithm emulating proton treatment planners' clinical practice is employed in post-processing to improve the output of the model. Forty-nine patients that received proton liver treatments between 2017 and 2020 were randomly divided into training (n = 31), validation (n = 7), and test sets (n = 11). AI-selected beam angles were compared with those angles selected by human planners, and the dosimetric outcome was investigated by creating plans using knowledge-based treatment planning.

RESULTS

For 7 of the 11 cases in the test set, AI-selected beam angles agreed with those chosen by human planners to within 20 degrees (median angle difference = 10°; mean = 18.6°). Moreover, out of the total 22 beam angles predicted by the model, 15 (68%) were within 10 degrees of the human-selected angles. The high correlation in beam angles resulted in comparable dosimetric statistics between proton treatment plans generated using AI- and human-selected angles. For the cases with beam angle differences exceeding 20°, the dosimetric analysis showed similar plan quality although with different emphases on organ-at-risk sparing.

CONCLUSIONS

This pilot study demonstrated the feasibility of a novel deep learning-based beam angle selection technique. Testing on liver cancer patients showed that the resulting plans were clinically viable with comparable dosimetric quality to those using human-selected beam angles. In tandem with auto-contouring and knowledge-based treatment planning tools, the proposed model could represent a pathway for nearly fully automated treatment planning in proton therapy. This article is protected by copyright. All rights reserved.

DockCoV2: a drug database against SARS-CoV-2

The current state of the COVID-19 pandemic is a global health crisis. To fight the novel coronavirus, one of the best-known ways is to block enzymes essential for virus replication. Currently, we know that the SARS-CoV-2 virus encodes about 29 proteins such as spike protein, 3C-like protease (3CLpro), RNA-dependent RNA polymerase (RdRp), Papain-like protease (PLpro), and nucleocapsid (N) protein. SARS-CoV-2 uses human angiotensin-converting enzyme 2 (ACE2) for viral entry and transmembrane serine protease family member II (TMPRSS2) for spike protein priming. Thus in order to speed up the discovery of potential drugs, we develop DockCoV2, a drug database for SARS-CoV-2. DockCoV2 focuses on predicting the binding affinity of FDA-approved and Taiwan National Health Insurance (NHI) drugs with the seven proteins mentioned above. This database contains a total of 3,109 drugs. DockCoV2 is easy to use and search against, is well cross-linked to external databases, and provides the state-of-the-art prediction results in one site. Users can download their drug-protein docking data of interest and examine additional drug-related information on DockCoV2. Furthermore, DockCoV2 provides experimental information to help users understand which drugs have already been reported to be effective against MERS or SARS-CoV. DockCoV2 is available at https://covirus.cc/drugs/.

Artificial Intelligence Aids Cardiac Image Quality Assessment for Improving Precision in Strain Measurements

OBJECTIVES

The aim of this study was to develop an artificial intelligence tool to assess echocardiographic image quality objectively.

BACKGROUND

Left ventricular global longitudinal strain (LVGLS) has recently been used to monitor cancer therapeutics-related cardiac dysfunction (CTRCD) but image quality limits its reliability.

METHODS

A DenseNet-121 convolutional neural network was developed for view identification from an athlete's echocardiographic dataset. To prove the concept that classification confidence (CC) can serve as a quality marker, values of longitudinal strain derived from feature tracking of cardiac magnetic resonance (CMR) imaging and strain analysis of echocardiography were compared. The CC was then applied to patients with breast cancer free from CTRCD to investigate the effects of image quality on the reliability of strain analysis.

RESULTS

CC of the apical 4-chamber view (A4C) was significantly correlated with the endocardial border delineation index. CC of A4C >900 significantly predicted a <15% relative difference in longitudinal strain between CMR feature tracking and automated echocardiographic analysis. Echocardiographic studies (n =752) of 102 patients with breast cancer without CTRCD were investigated. The strain analysis showed higher parallel forms, inter-rater, and test-retest reliabilities in patients with CC of A4C >900. During sequential comparisons of automated LVGLS in individual patients, those with CC of A4C >900 had a lower false positive detection rate of CTRCD.

CONCLUSIONS

CC of A4C was associated with the reliability of automated LVGLS and could also potentially be used as a filter to select comparable images from sequential echocardiographic studies in individual patients and reduce the false positive detection rate of CTRCD.