Transfer learning for fluence map prediction in adrenal stereotactic body radiation therapy

Objective: To design a deep transfer learning framework for modeling fluence map predictions for stereotactic body radiation therapy (SBRT) of adrenal cancer and similar sites that usually have a small number of cases. Approach: We developed a transfer learning framework for adrenal SBRT planning that leverages knowledge in a pancreas SBRT planning model. Treatment plans from the two sites had different dose prescriptions and beam settings but both prioritized gastrointestinal sparing. A base framework was first trained with 100 pancreas cases. This framework consists of two convolutional neural networks (CNN), which predict individual beam doses (BD-CNN) and fluence maps (FM-CNN) sequentially for 9-beam intensity-modulated radiation therapy (IMRT) plans. Forty-five adrenal plans were split into training/validation/test sets with the ratio of 20/10/15. The base BD-CNN was re-trained with transfer learning using 5/10/15/20 adrenal training cases to produce multiple candidate adrenal BD-CNN models. The base FM-CNN was directly used for adrenal cases. The deep learning (DL) plans were evaluated by several clinically relevant dosimetric endpoints, producing a percentage score relative to the clinical plans. Main results: Transfer learning significantly reduced the number of training cases and training time needed to train such a DL framework. The adrenal transfer learning model trained with 5/10/15/20 cases achieved validation plan scores of 85.4/91.2/90.7/89.4, suggesting that model performance saturated with 10 training cases. Meanwhile, a model using all 20 adrenal training cases without transfer learning only scored 80.5. For the final test set, the 5/10/15/20-case models achieved scores of 73.5/75.3/78.9/83.3. Significance: It is feasible to use deep transfer learning to train an IMRT fluence prediction framework. This technique could adapt to different dose prescriptions and beam configurations. This framework potentially enables DL modeling for clinical sites that have a limited dataset, either due to few cases or due to rapid technology evolution.

PQ Forum #23. PQ Documents List In Validation Plan

“PQ Forum” provides a mechanism for validation practitioners to share information related to Stage 2 Process Qualification in the validation lifecycle. Information about supporting activities such as design and development, equipment, and analytical validation will also be shared. The information provided should be helpful and practical so as to...

PQ Forum #22. Validation Plan Document Structure

A carefully written Validation Plan including thoughtful discussion in its respective sections ensures a logical and complete validation project. The following discussion proposes a simple and straightforward Validation Plan document structure that is applicable to all validation/qualification at a pharma manufacturing site.

How to Validate in silico Deployment of Coronary Stents: Strategies and Limitations in the Choice of Comparator

This study aims at proposing and discussing useful indications to all those who need to validate a numerical model of coronary stent deployment. The proof of the reliability of a numerical model is becoming of paramount importance in the era of in silico trials. Recently, the ASME V&V Standard Committee for medical devices prepared the V&V 40 standard document that provides a framework that guides users in establishing and assessing the relevance and adequacy of verification and validation activities performed for proving the credibility of models. To the knowledge of the authors, only a few examples of the application of the V&V 40 framework to medical devices are available in the literature, but none about stents. Specifically, in this study, the authors wish to emphasize the choice of a relevant set of experimental activities to provide data for the validation of computational models aiming to predict coronary stent deployment. Attention is focused on the use of ad hoc 3D-printed mock vessels in the validation plan, which could allow evaluating aspects of clinical relevance in a representative but controlled environment.

Meter-scale retrieval of industrial methane emissions using GHGSat’s satellite constellation

<p>Actionable feedback to industrial operators is extremely valuable to help them reduce their greenhouse gas emissions. With this goal in mind, GHGSat launched in 2016 a demonstration satellite called GHGSat-D (“Claire”). It was the first satellite built specifically to detect and quantify methane emissions from individual sites.</p><p>With the launches of GHGSat-C1 (“Iris”) in September 2020 and of GHGSat-C2 (“Hugo”) planned in January 2021, GHGSat will have three methane-sensing meter-scale resolution satellites in orbit. In addition to those satellites, GHGSat has also deployed an aircraft version of the instrument to survey specific areas with even lower detection threshold thanks to its higher spatial resolution.</p><p>This presentation will show the improvements done since GHGSat-D that allow our instruments to reach column precision of  1% of background. With this enhanced sensitivity, sources such as oil and gas facilities, mines, landfills and dams can be measured from space. Emission quantification of various sources will be presented and will demonstrate that GHGSat-C1 is approaching its target detection threshold of 100 kg/h. We will also illustrate the complementarity of GHGSat’s instruments with Sentinel-5P, the first ones able to detect individual sources with low emission rates, the second able to measure daily and with high accuracy global methane concentrations. We will also discuss the data calibration and validation plan of our instruments. Finally, an update on the future expansion of GHGSat’s constellation will be given.</p>

ASSESSING THE DRIVER'S RISK PERCEPTION DURING AUTONOMOUS DRIVING

Advanced Driver-Assistance Systems were created to address the driver's failures. All these ADAS are a part of the evolution of the vehicles towards whole automation. To complete its launch in the automotive market, autonomous vehicles have to pass safety tests to acquire the consumers’ trust. To receive the approval of the public, the self-driving car has to take into account the human feeling. The risk perceived by the driver is one of the new emotional form to integrate at the validation plan. The purpose of this study is to examine the perception of the risk of a self-driving car's driver.

Validation of the EarthCARE Mission

<p>This poster will address the geophysical validation for EarthCARE. This mission is developed by the European Space Agency (ESA) in cooperation with the Japan Aerospace eXploration Agency (JAXA); both space agencies also agreed to define and coordinate a joint EarthCARE Validation programme. Beside providing the Cloud Profile Radar instrument and making available the related ground processing facilities, JAXA is as well responsible for the commissioning of the CPR, including the associated Validation Plan and activities. ESA will then integrate the CPR Validation Plan part into the joint EarthCARE Scientific Validation Implementation Plan. The two Agencies have already begun to consolidate this joint Scientific Validation Implementation Plan, and its overall status will be presented. The poster will then focus on the ESA-led Validation activities, in particular on validation of the Level 1 products of the ESA instruments (ATLID, BBR, MSI) and on the ESA-developed Level 2 products. These ESA Validation activities have been the outcome of an announcement of opportunity that was issued in 2017 and for which more than 30 proposals had been received. A broad peer review of this programme took place in 2018 during the 1st ESA Validation Workshop in Bonn (held in concomitance with the 7th EarthCARE Science Workshop), and the conclusion was that if all Principal Investigators succeed to secure the corresponding funding, then the combined programme is adequate, with few areas for improvement remaining. Therefore, late opportunity still exists for supporting and complementing the EarthCARE Validation Plan</p>