Semi-supervised planning method for breast electronic tissue compensation treatments based on breast radius and separation Cover

.blurhash-client-img { display: none !important; }

Semi-supervised planning method for breast electronic tissue compensation treatments based on breast radius and separation

Radiology and Oncology

Volume 55 (2021): Issue 1 (March 2021)

By: Alexander R. Podgorsak and Lalith K. Kumaraswamy

Open Access

|Dec 2020

Figures & Tables

Left shows how the breast radius varies in the cranio-caudal direction, right shows how the breast separation varies in the cranio-caudal direction for the same collected patient breast.

Variation in the optimal transmission penetration depth (TPD) over all slices in the cranio-caudal direction compared with the TPD considering a constant TPD or a TPD using the three-region breast approach.

CT-simulation axial slice with fitted ellipse (red) and breast radius and separation estimation (yellow) and hand-measurement (blue) of left breast overlaid. Included is the location of the hand measurement for radius and separation in blue.

Correlation between breast separation and radius with the x-ray fluence needed to deliver homogenous dose distribution to the breast treatment volume using a 6 MV beam.

Optimal fluence maps from (A) assuming a single penetration depth, from (B) assuming a three-region breast model, and (C) our proposed model. Colorbar shows beam fluence.

(A, C, E) axial, coronal, and sagittal views of the breast treatment volume showing the isodose color washes from the treatment plan generated with the proposed algorithm. (B, D, F) axial, coronal, and sagittal views of the breast treatment volume showing the isodose color washes from the treatment plan generated by assuming a constant transmission penetration depth of 30%.

Summary of mean dose to the heart and ipsilateral lung V20 Gy for all ten collected courses_ Compared are the treatments generated with our proposed algorithm, plans created using the 3-region breast model proposed by Alghufaili, those plans optimized by a medical dosimetrist, and plans generated by assuming a single transmission penetration depth (TPD) within the treatment planning software

		Mean dose to the heart (cGy)				Ipsilateral lung V_{20 Gy} (%)

Number	Proposed Work	3-region breast model	Dosimetrist Optimized	Single TPD	Proposed Work	3-region breast model	Dosimetrist Optimized	Single TPD
1	11.0	15.0	13.7	13.4	0.42	2.30	2.41	2.05
2	298	296	294	327	18.2	18.3	16.7	20.3
3	177	165	109	160	5.12	4.91	3.56	4.65
4	271	267	149	250	9.09	9.03	8.07	7.93
5	40.8	44.9	39.7	42.0	16.6	16.8	15.8	16.2
6	113	112	108	111	8.95	8.84	8.33	8.59
7	47.0	52.4	47.7	51.7	9.56	11.2	9.9	10.9
8	26.9	28.9	26.5	28.7	11.4	11.2	11.4	10.9
9	95.9	128	89.5	125	9.18	10.8	7.8	10.6
10	406	402	159	397	18.7	18.5	12.4	18.2
Average	149	151	104	151	10.7	11.2	9.65	11.0

Summary of global dose maximum and clinical target volume (CTV) minimum values for all ten collected treatment courses_ Compared are the treatments generated with our proposed algorithm, plans created using the 3-region breast model proposed by Alghufaili, those plans optimized by a medical dosimetrist, and plans generated by assuming a single transmission penetration depth (TPD) within the treatment planning software

		Global dose max (%)				CTV dose min (%)

Number	Proposed work	3-region breast model	Dosimetrist optimized	Single TPD	Proposed work	3-region breast model	Dosimetrist optimized	Single TPD
1	107.7	108.5	109.1	113.3	96.9	97.3	104.1	96.0
2	109.7	111.3	106.5	111.8	74.4	75.0	76.9	74.2
3	107.4	108.1	107.5	112.7	95.5	88.7	95.3	89.1
4	107.5	110.4	105.9	109.8	95.5	95.2	95.0	96.8
5	107.7	109.7	104.0	105.0	95.1	95.9	98.1	87.8
6	107.5	108.7	105.5	111.4	97.9	96.0	99.3	97.4
7	108.0	114.2	105.9	116.3	95.4	95.4	95.7	95.4
8	107.6	112.6	106.6	114.6	95.3	92.8	97.8	94.1
9	107.5	115.5	106.0	118.1	95.6	95.6	95.1	95.0
10	107.5	108.6	107.8	114.4	86.1	81.6	87.8	87.2
Average	107.8	111.2	106.5	112.7	92.7	91.4	94.5	91.3

Agreement between breast radius and separation hand-measurements and automatic algorithm measurement in centimeter (cm) difference and percent difference

	Radius

Number	% Difference	Difference (cm)
1	9.6 [9.3–9.9]	0.70 [0.68–0.72]
2	6.2 [5.7–6.7]	0.40 [0.39–0.41]
3	22.0 [21.5–22.5]	1.00 [0.98–1.02]
Average	12.6 [10.4–14.8]	0.69 [0.59–0.79]

Summary of dose homogeneity indices (HI) for all ten collected treatment courses_ Compared are the treatments generated with our proposed algorithm, plans created using the 3-region breast model proposed by Alghufaili, those plans optimized by a medical dosimetrist, and plans generated by assuming a single transmission penetration depth (TPD) within the treatment planning software

Number	Proposed Work HI	3-region breast model	Dosimetrist Optimized HI	Single TPD HI
1	26.7	22.4	16.5	23.0
2	17.8	19.2	14.1	19.1
3	10.6	15.2	11.2	17.2
4	11.8	14.6	9.51	13.3
5	8.49	7.87	3.32	8.04
6	8.79	14.1	6.64	12.6
7	4.59	7.44	3.82	8.19
8	7.57	21.5	8.84	23.5
9	11.3	17.9	8.29	18.7
10	10.6	16.0	10.5	17.8
Average	12.6	15.6	9.87	17.0

DOI: https://doi.org/10.2478/raon-2020-0073 | Journal eISSN: 1581-3207 | Journal ISSN: 1318-2099

Journal RSS Feed

Language: English

Page range: 106 - 115

Submitted on: Jul 24, 2020

|

Accepted on: Oct 19, 2020

|

Published on: Dec 22, 2020

Published by: Association of Radiology and Oncology

In partnership with: Paradigm Publishing Services

Publication frequency: 4 issues per year

Keywords:

electronic compensation,

dose homogeneity index,

plan optimization

Related subjects:

Clinical medicine,

Internal medicine,

Haematology, oncology,

© 2020 Alexander R. Podgorsak, Lalith K. Kumaraswamy, published by Association of Radiology and Oncology
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Previous article Volume 55 (2021): Issue 1 (March 2021)Next article