Estimation of lung tumor position from multiple anatomical features on 4D‐<scp>CT</scp> using multiple regression analysis

Ono, Tomohiro; Nakamura, Mitsuhiro; Hirose, Yoshinori; Kitsuda, Kenji; Ono, Yuka; Ishigaki, Takashi; Hiraoka, Masahiro

doi:10.1002/acm2.12121

Cited by 6 publications

(12 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the MRA method, every value of the independent multiple predictor variables is associated with a dependent objective variable . The MRA equation is given by

y = a_{1} x_{1} + a_{2} x_{2} + a_{3} x_{3} + \dots + a_{p} x_{p} + c

where y is the predicted passing rate for each dosimetric accuracy, a i are partial regression coefficients ( i = 0, 1, 2, …, p ), x i are objective variables ( i = 0, 1, 2, …, p ), and c is a constant.…”

Section: Methodsmentioning

confidence: 99%

“…In the MRA method, every value of the independent multiple predictor variables is associated with a dependent objective variable. [27][28][29] The MRA equation is given by…”

Section: D2 Multiple Regression Analysismentioning

confidence: 99%

“…[24][25][26] MRA is a widely used linear analysis method that also provides fast prediction. [27][28][29] Both RTA and MRA can be considered as traditional machine learning methods. NNs comprise an artificial analysis method and are being widely used in medical physics.…”

Section: Introductionmentioning

confidence: 99%

“…RTA is a nonlinear analysis method and provides fast prediction and high interpretability of objective variables directly from the regression tree . MRA is a widely used linear analysis method that also provides fast prediction . Both RTA and MRA can be considered as traditional machine learning methods.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Prediction of dosimetric accuracy for VMAT plans using plan complexity parameters via machine learning

et al. 2019

Self Cite

View full text Add to dashboard Cite

Purpose The dosimetric accuracies of volumetric modulated arc therapy (VMAT) plans were predicted using plan complexity parameters via machine learning. Methods The dataset consisted of 600 cases of clinical VMAT plans from a single institution. The predictor variables (n = 28) for each plan included complexity parameters, machine type, and photon beam energy. Dosimetric measurements were performed using a helical diode array (ArcCHECK), and the dosimetric accuracy of the passing rates for a 5% dose difference (DD5%) and gamma index of 3%/3 mm (γ3%/3 mm) were predicted using three machine learning models: regression tree analysis (RTA), multiple regression analysis (MRA), and neural networks (NNs). First, the prediction models were applied to 500 cases of the VMAT plans. Then, the dosimetric accuracy was predicted using each model for the remaining 100 cases (evaluation dataset). The error between the predicted and measured passing rates was evaluated. Results For the 600 cases, the mean ± standard deviation of the measured passing rates was 92.3% ± 9.1% and 96.8% ± 3.1% for DD5% and γ3%/3 mm, respectively. For the evaluation dataset, the mean ± standard deviation of the prediction errors for DD5% and γ3%/3 mm was 0.5% ± 3.0% and 0.6% ± 2.4% for RTA, 0.0% ± 2.9% and 0.5% ± 2.4% for MRA, and –0.2% ± 2.7% and –0.2% ± 2.1% for NN, respectively. Conclusions NNs performed slightly better than RTA and MRA in terms of prediction error. These findings may contribute to increasing the efficiency of patient‐specific quality‐assurance procedures.

show abstract

“…In the MRA method, every value of the independent multiple predictor variables is associated with a dependent objective variable . The MRA equation is given by

y = a_{1} x_{1} + a_{2} x_{2} + a_{3} x_{3} + \dots + a_{p} x_{p} + c

Section: Methodsmentioning

confidence: 99%

“…In the MRA method, every value of the independent multiple predictor variables is associated with a dependent objective variable. [27][28][29] The MRA equation is given by…”

Section: D2 Multiple Regression Analysismentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Prediction of dosimetric accuracy for VMAT plans using plan complexity parameters via machine learning

et al. 2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…However, 4DCT can collect the motion information during the complete motion period. The reconstructed images can dynamically demonstrate the trajectory in the motion cycle, which can reflect the motion range of tumors in the three-dimensional directions (8).…”

Section: Introductionmentioning

confidence: 99%

Effects of respiratory motion on volumetric and positional difference of GTV in lung cancer based on 3DCT and 4DCT scanning

et al. 2018

View full text Add to dashboard Cite

Differences in gross target volume (GTV) and central point positions among moving lung cancer models constructed by CT scanning at different frequencies were compared, in order to explore the effect of different respiratory frequencies on the GTV constructions in moving lung tumors. Eight models in different shapes and sizes were established to stimulate lung tumors. The three-dimensional computed tomography (3DCT) and four-dimensional computed tomography (4DCT) scanning were performed at 10, 15 and 20 times/min in different models. Differences in GTV volumes and central point positions at different motion frequencies were compared by means of GTV3Ds (GTV3D-10, GTV3D-15, GTV3D-20) and IGTV4Ds (IGTV4D-10, IGTV4D-15, IGTV4D-20). Volumes of GTV3D-10, GTV3D-15, GTV3D-20 were 12.41±14.26, 10.38±11.18 and 12.50±15.23 cm3 respectively (P=0.687). Central point coordinates in the x-axis direction were −8.16±96.21, −8.57±96.08 and −8.56±95.73 respectively (P=0.968). Central point coordinates in the y-axis direction were 108.22±25.03, 110.41±22.47 and 109.04±24.24 (P=0.028). Central point coordinates in the z-axis direction were 65.19±13.68, 65.43±13.40 and 65.38±13.17 (P=0.902). The difference was significant in the y-axis direction (P=0.028). Volumes of IGTV4D-10, IGTV4D-15, IGTV4D-20 were 17.78±19.42, 17.43±19.56 and 17.44±18.80 cm3 (P=0.417). Central point coordinates in the x-axis direction were −7.73±95.93, −7.86±95.56 and −7.92±95.14 (P=0.325). Central point coordinates in the y-axis direction were 109.41±24.54, 109.60±24.13 and 109.16±24.28 (P=0.525). Central point coordinates in the z-axis direction were 65.52±13.31, 65.59±13.39 and 65.51±13.34 (P=0.093). However, the central point position of GTV in the head and foot direction by 3DCT scanning was severely affected by the respiratory frequency.

show abstract

Development of a prediction model for target positioning by using diaphragm waveforms extracted from CBCT projection images

Sakurai,

Ambo,

Nakamura

et al. 2023

J Applied Clin Med Phys

Self Cite

View full text Add to dashboard Cite

PurposeTo develop a prediction model (PM) for target positioning using diaphragm waveforms extracted from CBCT projection images.MethodsNineteen patients with lung cancer underwent orthogonal rotational kV x‐ray imaging lasting 70 s. IR markers placed on their abdominal surfaces and an implanted gold marker located nearest to the tumor were considered as external surrogates and the target, respectively. Four different types of regression‐based PM were trained using surrogate motions and target positions for the first 60 s, as follows: Scenario A: Based on the clinical scenario, 3D target positions extracted from projection images were used as they were (PMCL). Scenario B: The short‐arc 4D‐CBCT waveform exhibiting eight target positions was obtained by averaging the target positions in Scenario A. The waveform was repeated for 60 s (W4D‐CBCT) by adapting to the respiratory phase of the external surrogate. W4D‐CBCT was used as the target positions (PM4D‐CBCT). Scenario C: The Amsterdam Shroud (AS) signal, which depicted the diaphragm motion in the superior–inferior direction was extracted from the orthogonal projection images. The amplitude and phase of W4D‐CBCT were corrected based on the AS signal. The AS‐corrected W4D‐CBCT was used as the target positions (PMAS‐4D‐CBCT). Scenario D: The AS signal was extracted from single projection images. Other processes were the same as in Scenario C. The prediction errors were calculated for the remaining 10 s.ResultsThe 3D prediction error within 3 mm was 77.3% for PM4D‐CBCT, which was 12.8% lower than that for PMCL. Using the diaphragm waveforms, the percentage of errors within 3 mm improved by approximately 7% to 84.0%‐85.3% for PMAS‐4D‐CBCT in Scenarios C and D, respectively. Statistically significant differences were observed between the prediction errors of PM4D‐CBCT and PMAS‐4D‐CBCT.ConclusionPMAS‐4D‐CBCT outperformed PM4D‐CBCT, proving the efficacy of the AS signal‐based correction. PMAS‐4D‐CBCT would make it possible to predict target positions from 4D‐CBCT images without gold markers.

show abstract

Estimation of lung tumor position from multiple anatomical features on 4D‐CT using multiple regression analysis

Cited by 6 publications

References 24 publications

Prediction of dosimetric accuracy for VMAT plans using plan complexity parameters via machine learning

Prediction of dosimetric accuracy for VMAT plans using plan complexity parameters via machine learning

Effects of respiratory motion on volumetric and positional difference of GTV in lung cancer based on 3DCT and 4DCT scanning

Development of a prediction model for target positioning by using diaphragm waveforms extracted from CBCT projection images

Contact Info

Product

Resources

About