國立虎尾科技大學 |

A Deep Learning Model for V50%, V60%, and V66.7% Prediction in LINAC-Based Treatment Planning of Single-Iso-Multiple-Targets (SIMT) Stereotactic Radiosurgery (SRS).

紀錄類型:	書目-語言資料,手稿 : Monograph/item
正題名/作者:	A Deep Learning Model for V50%, V60%, and V66.7% Prediction in LINAC-Based Treatment Planning of Single-Iso-Multiple-Targets (SIMT) Stereotactic Radiosurgery (SRS)./
作者:	Khazaieli, Mercedeh.
面頁冊數:	1 online resource (48 pages)
附註:	Source: Masters Abstracts International, Volume: 84-11.
Contained By:	Masters Abstracts International84-11.
標題:	Medicine. -
電子資源:	click for full text (PQDT)
ISBN:	9798379571412

A Deep Learning Model for V50%, V60%, and V66.7% Prediction in LINAC-Based Treatment Planning of Single-Iso-Multiple-Targets (SIMT) Stereotactic Radiosurgery (SRS).
Khazaieli, Mercedeh.

A Deep Learning Model for V50%, V60%, and V66.7% Prediction in LINAC-Based Treatment Planning of Single-Iso-Multiple-Targets (SIMT) Stereotactic Radiosurgery (SRS). - 1 online resource (48 pages)

Source: Masters Abstracts International, Volume: 84-11.

Thesis (M.S.)--Duke University, 2023.

Includes bibliographical references

Brain metastases are a common complication of several types of cancer, with about 30-40% of cancer patients developing brain metastases during the course of their treatment. One treatment approach for brain metastases is LINAC-based single-isocenter multiple-target (SIMT) stereotactic radiosurgery (SRS). However, generating highquality SIMT SRS treatment plans can be time-consuming and can lead to large variations in plan quality, which can affect patient outcomes. Therefore, predicting achievable dosimetric outcomes can be valuable for reducing plan quality variation and improving planning efficiency. The amount of normal brain tissue receiving 12Gy during a single treatment is a known predictor of brain tissue toxicity, also known as radionecrosis. If the treatment involves 20Gy x 1fx, then the volume of normal brain tissue receiving 60% (V60%) of the treatment dose is equal to the volume receiving 12Gy. Similarly, if the treatment involves 18Gy x 1fx, then the volume of normal brain tissue receiving 66.7% (V66.7%) of the treatment dose is equal to the volume receiving 12Gy. The volume of normal brain tissue receiving 50% (V50%) of the treatment dose is a good measure of the gradient of the dose fall-off from the targets. A prediction of achievable V50% (gradient measure), V60% (assuming 20Gy x 1fx), and V66.7% (assuming 18Gy x 1fx) can assist physicians in the determination of fractionation schemes (i.e., single fx vs. multiple fx). Such predictions can be used as guidelines for planners to generate a SIMT plan more rapidly with reduced dosimetric variability. To this end, we developed deep learning networks for the prediction of V50%, V60%, and V66.7% based on each patient's target contours. A key technical innovation of this work is the spherical projection design: by projecting target contours on a spherical surface, the target contours in 3D collapse to a polar-azimuthal angular distribution map. This transformation is view-invariant and enables a dimensional reduction for deep learning input without losing volumetric information. Our results indicate promising potential but there is a need for further work to improve the accuracy of our predictions.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2024

Mode of access: World Wide Web

ISBN: 9798379571412Subjects--Topical Terms:

644133
Medicine.
Subjects--Index Terms:

Deep learning modelIndex Terms--Genre/Form:

554714
Electronic books.

A Deep Learning Model for V50%, V60%, and V66.7% Prediction in LINAC-Based Treatment Planning of Single-Iso-Multiple-Targets (SIMT) Stereotactic Radiosurgery (SRS).
LDR:03632ntm a22003977 4500 001 1147094
005 20240909062213.5
006 m o d
007 cr bn ---uuuuu
008 250605s2023 xx obm 000 0 eng d
020 $a 9798379571412
035 $a (MiAaPQ)AAI30312420
035 $a AAI30312420
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Khazaieli, Mercedeh. $3 1472724
245 1 2 $a A Deep Learning Model for V50%, V60%, and V66.7% Prediction in LINAC-Based Treatment Planning of Single-Iso-Multiple-Targets (SIMT) Stereotactic Radiosurgery (SRS).
264 0 $c 2023
300 $a 1 online resource (48 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: Masters Abstracts International, Volume: 84-11.
500 $a Advisor: Wang, Chunhao.
502 $a Thesis (M.S.)--Duke University, 2023.
504 $a Includes bibliographical references
520 $a Brain metastases are a common complication of several types of cancer, with about 30-40% of cancer patients developing brain metastases during the course of their treatment. One treatment approach for brain metastases is LINAC-based single-isocenter multiple-target (SIMT) stereotactic radiosurgery (SRS). However, generating highquality SIMT SRS treatment plans can be time-consuming and can lead to large variations in plan quality, which can affect patient outcomes. Therefore, predicting achievable dosimetric outcomes can be valuable for reducing plan quality variation and improving planning efficiency. The amount of normal brain tissue receiving 12Gy during a single treatment is a known predictor of brain tissue toxicity, also known as radionecrosis. If the treatment involves 20Gy x 1fx, then the volume of normal brain tissue receiving 60% (V60%) of the treatment dose is equal to the volume receiving 12Gy. Similarly, if the treatment involves 18Gy x 1fx, then the volume of normal brain tissue receiving 66.7% (V66.7%) of the treatment dose is equal to the volume receiving 12Gy. The volume of normal brain tissue receiving 50% (V50%) of the treatment dose is a good measure of the gradient of the dose fall-off from the targets. A prediction of achievable V50% (gradient measure), V60% (assuming 20Gy x 1fx), and V66.7% (assuming 18Gy x 1fx) can assist physicians in the determination of fractionation schemes (i.e., single fx vs. multiple fx). Such predictions can be used as guidelines for planners to generate a SIMT plan more rapidly with reduced dosimetric variability. To this end, we developed deep learning networks for the prediction of V50%, V60%, and V66.7% based on each patient's target contours. A key technical innovation of this work is the spherical projection design: by projecting target contours on a spherical surface, the target contours in 3D collapse to a polar-azimuthal angular distribution map. This transformation is view-invariant and enables a dimensional reduction for deep learning input without losing volumetric information. Our results indicate promising potential but there is a need for further work to improve the accuracy of our predictions.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2024
538 $a Mode of access: World Wide Web
650 4 $a Medicine. $3 644133
650 4 $a Mathematics. $3 527692
650 4 $a Physics. $3 564049
650 4 $a Neurosciences. $3 593561
653 $a Deep learning model
653 $a Stereotactic radiosurgery
653 $a Brain metastases
653 $a Primary systemic tumors
655 7 $a Electronic books. $2 local $3 554714
690 $a 0564
690 $a 0405
690 $a 0605
690 $a 0317
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a Duke University. $b Medical Physics. $3 1472725
773 0 $t Masters Abstracts International $g 84-11.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30312420 $z click for full text (PQDT)