東華大學圖書館 |

Enhancements to the Scatter Display System (SDS) for Staff Dose Management During Fluoroscopically Guided Interventional Procedures /

Record Type:	Electronic resources : Monograph/item
Title/Author:	Enhancements to the Scatter Display System (SDS) for Staff Dose Management During Fluoroscopically Guided Interventional Procedures // Martina Precious Orji.
Author:	Orji, Martina Precious,
Description:	1 electronic resource (143 pages)
Notes:	Source: Dissertations Abstracts International, Volume: 86-12, Section: B.
Contained By:	Dissertations Abstracts International86-12B.
Subject:	Medical imaging. -
Online resource:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=32043099
ISBN:	9798280777248

Enhancements to the Scatter Display System (SDS) for Staff Dose Management During Fluoroscopically Guided Interventional Procedures /
Orji, Martina Precious,

Enhancements to the Scatter Display System (SDS) for Staff Dose Management During Fluoroscopically Guided Interventional Procedures /Martina Precious Orji. - 1 electronic resource (143 pages)

Source: Dissertations Abstracts International, Volume: 86-12, Section: B.

Radiation exposure from interventional fluoroscopy procedures creates concern in medical procedures, especially due to the enormous scatter radiation that affects both patients and healthcare staff resulting in radiation-induced health hazards that include skin burn injury and eye-lens cataract formation as well as increased cancer risk. The occupational dose of the personnel involved in this type of procedure is seen to be the highest among others, making radiation safety measures a critical component in the workflow for fluoroscopically guided medical procedures.To address this, a real-time Scattered Radiation Display System (SDS) was developed by our group. This system displays a color-coded mapping of procedure-room scatter distributions by accessing the digital output updates of technique and geometric parameters from the angiography system through the controller area network (CAN) bus device. The system provides prompt visual feedback to the staff by display of the procedure-room scatter radiation distribution in horizontal planes at selectable heights, providing an awareness of regions in the room with high radiation levels. This dissertation describes enhancements made to the SDS improving and fostering a safer clinical environment during procedures. The system was initially developed for scatter dose distributions in a limited area near the patient table, and with procedures limited to thoracic and cranial regions of the patient. However, this existing setup posed limitations. The small virtual room size did not accurately reflect standard procedure room dimensions and could not provide sufficient feedback of scatter levels for all medical personnel as they moved about in the room. Additionally, the SDS uses a Lookup Table (LUT) to display scatter distribution, which is limited by its high data storage and load time requirements.To address these challenges, the SDS framework was expanded by creating a robust database of scatter dose distributions using Monte Carlo simulations. These simulations incorporated Zubal phantoms for various anatomical regions including the head, neck, chest, abdomen, and legs. To preprocess the phantom data, create scatter detection plane and extract required material and density information, MATLAB tools were used. The extracted materials and associated densities were then converted into a compatible format for input into the DOSXYZnrc GUI for dose mapping simulation. To accurately represent a typical clinical scenario, the simulation volume was increased, with matrix dimensions ranging from 450x450x76 to 650x650x76 and 1250x1250x76 with a resolution of 4 mm per pixel. This simulation were done for different geometric and exposure parameters.The DNN approach was further improved to efficiently learn scatter dose distribution patterns from the Monte Carlo generated data. Instead of running thousands of new simulations to train the network at higher resolutions, an approach involving linear interpolation of existing matrices was used. This approach generated additional training data while significantly reducing the computational burden and saving time. Other evaluated training set parameter changes included variation of distribution spatial resolution, the dropout of data near the isocenter and logarithmic data compression. When compared with ground truth, network predictions, using an expanded dataset obtained via linear interpolation provided superior accuracy with an average DNN MAPE which is below 10% for scattered distributions across all gantry angles in the test set. There is no significant difference in accuracy with downsampling of the data. However the downsampling method reduces the network size, which helps to reduce the storage requirements, and the average prediction time is reduced from about 0.2s to 0.1s, which helps maintain real-time feedback by enabling a continuous update of the distribution throughout the procedure without causing significant delay. On the other hand, dataset value range reduction whether by logarithmic conversion or isocenter dropout of data, did not show a significant improvement in average MAPE values. The same approach but with variation in a few hyperparameters, was adopted to train separate DNNs for the prediction accuracy of the 1250x1250x76 matrix of the expanded procedure room for each body region imaging at height approximately to waist level (120 cm from floor). Different networks that are capable of making prediction of large procedure-room scatter dose distributions during fluoroscopic guided procedures in the head, neck, chest, abdomen and leg body regions, were developed. When the prediction is compared with the MC ground truth, the mean absolute percent error (MAPE) < 10% for head, neck and leg imaging and <15% for chest and abdomen imaging are observed. While the mean absolute error (MAE) for the same expanded room scatter distribution prediction accuracy of all body region imaging is <3mGy/Gy on average. Another enhancement made to the SDS system was that the trained DNNs were further integrated into the system using MATLAB tools and frameworks for SDS GUI. The SDS GUI layout was modified to accommodate callback functions for fields and buttons that will add input data, process the input data, run it through the model, and return, then make predictions.An additional database was created consisting of Monte Carlo calculated data of scatter dose distribution using an ROI attenuator for dose reduction. A DNN model was trained using this database, resulting in a prediction accuracy of about 5%. This was developed to analyze scatter dose reductions in the procedure room using region-of-interest (ROI) attenuators. In the procedure room scatter measurement using 150 cc Keithley ionization chamber and PTW electrometer, for a staff standing at 50 cm from the table centerline, an overall dose reduction at eye level was shown to be about 50% for head imaging and 70% for chest and abdomen imaging at 80 kVp for all gantry angles when using a ROI attenuator with about 20% beam transmission. The dose reduction is reduced with an increase in kVp for the same attenuator as the transmission factor increases and is nearly independent of CRA/CAU angle. In general, the use of a ROI attenuator can provide a high reduction of scattered radiation to the interventional room staff without limiting the displayed FOV.Overall, these enhancements make the SDS a more robust system for managing occupational radiation exposure by providing comprehensive, real-time feedback, enabling staff to recognize and manage the high-dose areas more effectively, and contributing to safer procedural environments.

English

ISBN: 9798280777248Subjects--Topical Terms:

3172799
Medical imaging.
Subjects--Index Terms:

Deep neural network

Enhancements to the Scatter Display System (SDS) for Staff Dose Management During Fluoroscopically Guided Interventional Procedures /
LDR:08268nmm a22004573i 4500 001 2434912
005 20251001062554.5
006 m o d
007 cr|nu||||||||
008 261011s2025 miu||||||m |||||||eng d
020 $a 9798280777248
035 $a (MiAaPQD)AAI32043099
035 $a AAI32043099
040 $a MiAaPQD $b eng $c MiAaPQD $e rda
100 1 $a Orji, Martina Precious, $e author. $0 (orcid)0009-0007-8306-4876 $3 3825057
245 1 0 $a Enhancements to the Scatter Display System (SDS) for Staff Dose Management During Fluoroscopically Guided Interventional Procedures / $c Martina Precious Orji.
264 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2025
300 $a 1 electronic resource (143 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertations Abstracts International, Volume: 86-12, Section: B.
500 $a Advisors: Bednarek, Daniel R. Committee members: Rudin, Stephen; Ionita, Ciprian; Yao, Rutao.
502 $b Ph.D. $c State University of New York at Buffalo $d 2025.
520 $a Radiation exposure from interventional fluoroscopy procedures creates concern in medical procedures, especially due to the enormous scatter radiation that affects both patients and healthcare staff resulting in radiation-induced health hazards that include skin burn injury and eye-lens cataract formation as well as increased cancer risk. The occupational dose of the personnel involved in this type of procedure is seen to be the highest among others, making radiation safety measures a critical component in the workflow for fluoroscopically guided medical procedures.To address this, a real-time Scattered Radiation Display System (SDS) was developed by our group. This system displays a color-coded mapping of procedure-room scatter distributions by accessing the digital output updates of technique and geometric parameters from the angiography system through the controller area network (CAN) bus device. The system provides prompt visual feedback to the staff by display of the procedure-room scatter radiation distribution in horizontal planes at selectable heights, providing an awareness of regions in the room with high radiation levels. This dissertation describes enhancements made to the SDS improving and fostering a safer clinical environment during procedures. The system was initially developed for scatter dose distributions in a limited area near the patient table, and with procedures limited to thoracic and cranial regions of the patient. However, this existing setup posed limitations. The small virtual room size did not accurately reflect standard procedure room dimensions and could not provide sufficient feedback of scatter levels for all medical personnel as they moved about in the room. Additionally, the SDS uses a Lookup Table (LUT) to display scatter distribution, which is limited by its high data storage and load time requirements.To address these challenges, the SDS framework was expanded by creating a robust database of scatter dose distributions using Monte Carlo simulations. These simulations incorporated Zubal phantoms for various anatomical regions including the head, neck, chest, abdomen, and legs. To preprocess the phantom data, create scatter detection plane and extract required material and density information, MATLAB tools were used. The extracted materials and associated densities were then converted into a compatible format for input into the DOSXYZnrc GUI for dose mapping simulation. To accurately represent a typical clinical scenario, the simulation volume was increased, with matrix dimensions ranging from 450x450x76 to 650x650x76 and 1250x1250x76 with a resolution of 4 mm per pixel. This simulation were done for different geometric and exposure parameters.The DNN approach was further improved to efficiently learn scatter dose distribution patterns from the Monte Carlo generated data. Instead of running thousands of new simulations to train the network at higher resolutions, an approach involving linear interpolation of existing matrices was used. This approach generated additional training data while significantly reducing the computational burden and saving time. Other evaluated training set parameter changes included variation of distribution spatial resolution, the dropout of data near the isocenter and logarithmic data compression. When compared with ground truth, network predictions, using an expanded dataset obtained via linear interpolation provided superior accuracy with an average DNN MAPE which is below 10% for scattered distributions across all gantry angles in the test set. There is no significant difference in accuracy with downsampling of the data. However the downsampling method reduces the network size, which helps to reduce the storage requirements, and the average prediction time is reduced from about 0.2s to 0.1s, which helps maintain real-time feedback by enabling a continuous update of the distribution throughout the procedure without causing significant delay. On the other hand, dataset value range reduction whether by logarithmic conversion or isocenter dropout of data, did not show a significant improvement in average MAPE values. The same approach but with variation in a few hyperparameters, was adopted to train separate DNNs for the prediction accuracy of the 1250x1250x76 matrix of the expanded procedure room for each body region imaging at height approximately to waist level (120 cm from floor). Different networks that are capable of making prediction of large procedure-room scatter dose distributions during fluoroscopic guided procedures in the head, neck, chest, abdomen and leg body regions, were developed. When the prediction is compared with the MC ground truth, the mean absolute percent error (MAPE) < 10% for head, neck and leg imaging and <15% for chest and abdomen imaging are observed. While the mean absolute error (MAE) for the same expanded room scatter distribution prediction accuracy of all body region imaging is <3mGy/Gy on average. Another enhancement made to the SDS system was that the trained DNNs were further integrated into the system using MATLAB tools and frameworks for SDS GUI. The SDS GUI layout was modified to accommodate callback functions for fields and buttons that will add input data, process the input data, run it through the model, and return, then make predictions.An additional database was created consisting of Monte Carlo calculated data of scatter dose distribution using an ROI attenuator for dose reduction. A DNN model was trained using this database, resulting in a prediction accuracy of about 5%. This was developed to analyze scatter dose reductions in the procedure room using region-of-interest (ROI) attenuators. In the procedure room scatter measurement using 150 cc Keithley ionization chamber and PTW electrometer, for a staff standing at 50 cm from the table centerline, an overall dose reduction at eye level was shown to be about 50% for head imaging and 70% for chest and abdomen imaging at 80 kVp for all gantry angles when using a ROI attenuator with about 20% beam transmission. The dose reduction is reduced with an increase in kVp for the same attenuator as the transmission factor increases and is nearly independent of CRA/CAU angle. In general, the use of a ROI attenuator can provide a high reduction of scattered radiation to the interventional room staff without limiting the displayed FOV.Overall, these enhancements make the SDS a more robust system for managing occupational radiation exposure by providing comprehensive, real-time feedback, enabling staff to recognize and manage the high-dose areas more effectively, and contributing to safer procedural environments.
546 $a English
590 $a School code: 0656
650 4 $a Medical imaging. $3 3172799
650 4 $a Biomedical engineering. $3 535387
650 4 $a Medicine. $3 641104
650 4 $a Physics. $3 516296
653 $a Deep neural network
653 $a Dose prediction
653 $a Monte Carlo simulations
653 $a Occupational dose
653 $a Radiation safety
653 $a ROI imaging
690 $a 0574
690 $a 0541
690 $a 0564
690 $a 0605
710 2 $a State University of New York at Buffalo. $b Medical Physics - Radiology. $e degree granting institution. $3 3825058
720 1 $a Bednarek, Daniel R. $e degree supervisor.
773 0 $t Dissertations Abstracts International $g 86-12B.
790 $a 0656
791 $a Ph.D.
792 $a 2025
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=32043099