Yusuke Anetai
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Yusuke Anetai has a total of 35 co-authors. Here are the top co-authors by number of shared publications:
- Hideki Takegawa
Kansai Medical University
9 shared publications - Yuhei Koike
Department of Radiology, Kansai Medical University , 2-5-1 Shinmachi, Hirakata, Osaka, 573-1010
7 shared publications - Satoaki Nakamura
Division of Radiation Oncology, Kansai Medical University , 2-5-1 Shinmachi, Hirakata, Osaka, 573-1010
6 shared publications - Kentaro Doi
AI Center for Health and Biomedical Research (ArCHER), National Institutes of Biomedical Innovation, Health and Nutrition, Osaka, 566-0002, Japan
5 shared publications Yuhei Koike
Department of Radiology, Kansai Medical University, Osaka, Japan
4 shared publicationsKen Yoshida
Radiology, Kansai Medical University, 2-5-1, Shinmachi, Hirakata, Osaka, 573-1010, Japan
3 shared publications
Latest Publications
- Open Access
- 23 February 2026
- Published by IOP Publishing in Physics in Medicine & Biology
- Vol. 71 (4) , 045015
Practical gEUD optimization technique for stereotactic radiation therapy based on a theoretical reinterpretation using the gEUD curve concept
Objective. The generalized equivalent uniform dose (gEUD) is a well-established metric for radiotherapy dose optimization, particularly for normal tissues. However, the lack of theoretical clarity in its application has often led to empirical use in clinical practice. This study aims to reformulate gEUD-based optimization in a theoretical framework using the gEUD curve concept, and to develop a robust optimization strategy tailored for stereotactic radiation therapy (SRT), with a specific focus on bone and spinal metastases. Approach. We interpreted the gEUD as a smooth function of its a -parameter, forming a continuous curve whose deformability decreases with increasing a . Based on this understanding, we proposed two methods: the multiple gEUD objective (MgEUDO) for optimizing the remaining volume at risk, and the selective gEUD objective (SgEUDO) for critical organs at risk. These methods were retrospectively evaluated in 22 patients who received five-fraction SRT (35 Gy total prescription). Main results. Using consistent gEUD-based optimization, all cases achieved clinically favorable dose distributions. Compared to conventional normal tissue objective constraints, the proposed strategy reduced mean dose to surrounding tissues by 10%, while improving tumor dose coverage by 0.6%. SgEUDO further achieved 39% and 37.5% mean dose reductions in the left and right kidneys, respectively. Significance. Our theoretical and practical refinement of gEUD optimization enables systematic control of dose distribution with reduced inter-planner variability. The combined MgEUDO and SgEUDO strategies provide a generalizable and clinically effective framework for high-precision radiotherapy.
- 23 February 2026
- Published by IOP Publishing in Physics in Medicine & Biology
- Vol. 71 (4) , 045015
Open AccessPractical gEUD optimization technique for stereotactic radiation therapy based on a theoretical reinterpretation using the gEUD curve concept
Objective. The generalized equivalent uniform dose (gEUD) is a well-established metric for radiotherapy dose optimization, particularly for normal tissues. However, the lack of theoretical clarity in its application has often led to empirical use in clinical practice. This study aims to reformulate gEUD-based optimization in a theoretical framework using the gEUD curve concept, and to develop a robust optimization strategy tailored for stereotactic radiation therapy (SRT), with a specific focus on bone and spinal metastases. Approach. We interpreted the gEUD as a smooth function of its a -parameter, forming a continuous curve whose deformability decreases with increasing a . Based on this understanding, we proposed two methods: the multiple gEUD objective (MgEUDO) for optimizing the remaining volume at risk, and the selective gEUD objective (SgEUDO) for critical organs at risk. These methods were retrospectively evaluated in 22 patients who received five-fraction SRT (35 Gy total prescription). Main results. Using consistent gEUD-based optimization, all cases achieved clinically favorable dose distributions. Compared to conventional normal tissue objective constraints, the proposed strategy reduced mean dose to surrounding tissues by 10%, while improving tumor dose coverage by 0.6%. SgEUDO further achieved 39% and 37.5% mean dose reductions in the left and right kidneys, respectively. Significance. Our theoretical and practical refinement of gEUD optimization enables systematic control of dose distribution with reduced inter-planner variability. The combined MgEUDO and SgEUDO strategies provide a generalizable and clinically effective framework for high-precision radiotherapy.
- Open Access
- 28 May 2025
- Published by Springer Nature in International Journal of Computer Assisted Radiology and Surgery
- Vol. 20 (8) , 1-9
Homology-feature-assisted quantification of fibrotic lesions in computed tomography images: a proof of concept for CT image feature-based prediction for gene-expression-distribution
Purpose: Computed tomography (CT) image is promising for diagnosing of interstitial idiopathic pneumonias (IIPs); however, quantification of IIPs lesions in CT images is required. This study aimed to quantitatively evaluate fibrotic lesions in CT images using homology-based image analysis. Methods: We collected publicly available CT images comprising 47 fibrotic images and 36 non-fibrotic images. The homology-profile (HP) image analysis method provides b0 and b1 profiles, indicating the number of isolated components and holes in a binary image. We locally applied the HP method to the CT image and generated homology-based feature (HF) maps as resultant images. The collected images were randomly divided into the tuning dataset and the testing dataset. The cut-off value for classifying the HF map for fibrotic or non-fibrotic images was defined using receiver operating characteristic (ROC) analysis with the tuning dataset. This cut-off value was evaluated using the testing dataset with accuracy, sensitivity, specificity, and precision. Results: We successfully visualized the quantification of fibrotic lesions in the HF map. The b0 HF map was more suitable for quantifying fibrotic lesions than b1. The mean cut-off value of the b0 HF map was 199, with all performances achieved at 1.0. Furthermore, the classification of the b0 HF map for fibrotic or lung cancer images achieved all maximum performances at 1.0. Conclusion: This study demonstrated the feasibility of using the HF in quantitatively evaluating fibrotic lesions in CT images. Our proposed HP-based method can also be promising in quantifying the fibrotic lesions of patients with IIPs, which can be applicable to assist the diagnosis of IIPs.
- 28 May 2025
- Published by Springer Nature in International Journal of Computer Assisted Radiology and Surgery
- Vol. 20 (8) , 1-9
Open AccessHomology-feature-assisted quantification of fibrotic lesions in computed tomography images: a proof of concept for CT image feature-based prediction for gene-expression-distribution
Purpose: Computed tomography (CT) image is promising for diagnosing of interstitial idiopathic pneumonias (IIPs); however, quantification of IIPs lesions in CT images is required. This study aimed to quantitatively evaluate fibrotic lesions in CT images using homology-based image analysis. Methods: We collected publicly available CT images comprising 47 fibrotic images and 36 non-fibrotic images. The homology-profile (HP) image analysis method provides b0 and b1 profiles, indicating the number of isolated components and holes in a binary image. We locally applied the HP method to the CT image and generated homology-based feature (HF) maps as resultant images. The collected images were randomly divided into the tuning dataset and the testing dataset. The cut-off value for classifying the HF map for fibrotic or non-fibrotic images was defined using receiver operating characteristic (ROC) analysis with the tuning dataset. This cut-off value was evaluated using the testing dataset with accuracy, sensitivity, specificity, and precision. Results: We successfully visualized the quantification of fibrotic lesions in the HF map. The b0 HF map was more suitable for quantifying fibrotic lesions than b1. The mean cut-off value of the b0 HF map was 199, with all performances achieved at 1.0. Furthermore, the classification of the b0 HF map for fibrotic or lung cancer images achieved all maximum performances at 1.0. Conclusion: This study demonstrated the feasibility of using the HF in quantitatively evaluating fibrotic lesions in CT images. Our proposed HP-based method can also be promising in quantifying the fibrotic lesions of patients with IIPs, which can be applicable to assist the diagnosis of IIPs.
- 6 May 2025
- Published by Springer Nature in Physical and Engineering Sciences in Medicine
- Vol. 48 (2) , 797-811
Visualization and evaluation of the quality variations of EBT4 Gafchromic film using multidimensional scaling and Lie derivative image analysis
Film-specific uniformity variations in packages are known to significantly diminish the effectiveness of the one-scan protocol, a commonly used film dosimetry method. This method universally adopts the reference dose–response with rescaling linearly from the relationship of the known dose and the unexposed state. This study aims to visualize and quantify the variation in unexposed film-specific uniformity in a package to evaluate the suitability of the reference dose response using machine-learning method. Fourteen EBT4 films (#00–#13) were selected from two lot packages. Nine grid-spaced 100 × 100 pixel (72 dpi) patches were obtained from the color images of EBT4 film sheet using a single scanner with landscape (scan A) and portrait (scan B) scan orientations. The reference patch was set at the center of film #00. For this study, multidimensional scaling (MDS) and Lie derivative image analysis (LDIA) were applied to the patch data with respect to the red (R)/green (G)/blue (B) channels. MDS is a suitable method for analyzing non-linear data with similarity, which provides a map of data objects according to a distance metric. LDIA directly detects the deviation vector field between image gradients. The film-specific uniformity was measured at 1/10000 scaled pixel value as a scalar distribution. The image flow field was obtained as a negative gradient of the scalar distribution. Two similarity metrics were defined for comparison with the reference patch: (1) MDSr (the distance parameter in the MDS map from the origin) and (2) Stot (summed S-value in each patch, where S-value represents the vorticity of the deviation vector field obtained via the Lie derivative). MDSr highly correlated with the absolute pixel value difference from the reference patch except for the blue channel in which a favorable package was detected for the reference dose response. Stot quantified the film-uniformity variation from the reference, independent of the dataset, and detected the unfavorable film state as Stot < 0.8 in the blue channel. We visualized and quantified the variation in film-specific uniformity in a lot package using MDS and LDIA, thereby quantitatively determining the unfavorable condition for applying the reference dose–response.
- 6 May 2025
- Published by Springer Nature in Physical and Engineering Sciences in Medicine
- Vol. 48 (2) , 797-811
Visualization and evaluation of the quality variations of EBT4 Gafchromic film using multidimensional scaling and Lie derivative image analysis
Film-specific uniformity variations in packages are known to significantly diminish the effectiveness of the one-scan protocol, a commonly used film dosimetry method. This method universally adopts the reference dose–response with rescaling linearly from the relationship of the known dose and the unexposed state. This study aims to visualize and quantify the variation in unexposed film-specific uniformity in a package to evaluate the suitability of the reference dose response using machine-learning method. Fourteen EBT4 films (#00–#13) were selected from two lot packages. Nine grid-spaced 100 × 100 pixel (72 dpi) patches were obtained from the color images of EBT4 film sheet using a single scanner with landscape (scan A) and portrait (scan B) scan orientations. The reference patch was set at the center of film #00. For this study, multidimensional scaling (MDS) and Lie derivative image analysis (LDIA) were applied to the patch data with respect to the red (R)/green (G)/blue (B) channels. MDS is a suitable method for analyzing non-linear data with similarity, which provides a map of data objects according to a distance metric. LDIA directly detects the deviation vector field between image gradients. The film-specific uniformity was measured at 1/10000 scaled pixel value as a scalar distribution. The image flow field was obtained as a negative gradient of the scalar distribution. Two similarity metrics were defined for comparison with the reference patch: (1) MDSr (the distance parameter in the MDS map from the origin) and (2) Stot (summed S-value in each patch, where S-value represents the vorticity of the deviation vector field obtained via the Lie derivative). MDSr highly correlated with the absolute pixel value difference from the reference patch except for the blue channel in which a favorable package was detected for the reference dose response. Stot quantified the film-uniformity variation from the reference, independent of the dataset, and detected the unfavorable film state as Stot < 0.8 in the blue channel. We visualized and quantified the variation in film-specific uniformity in a lot package using MDS and LDIA, thereby quantitatively determining the unfavorable condition for applying the reference dose–response.
- Open Access
- 6 December 2024
- Published by IOP Publishing in Physics in Medicine & Biology
- Vol. 69 (24) , 245007
Diffusion equation quantification: selective enhancement algorithm for bone metastasis lesions in CT images
Objective. Diffusion equation (DE) imaging processing is promising to enhance images showing lesions of bone metastasis (LBM). The Perona–Malik diffusion (PMD) model, which has been widely used and studied, is an anisotropic diffusion processing method to denoise or extract objects from an image effectively. However, the smoothing characteristics of PMD or its related method hinder extraction and enhancement of soft tissue regions of medical image such as computed tomography (CT), typically leaving an indistinct region with ambient tissues. Moreover, PMD expands the border region of the objects. A novel diffusion methodology must be used to enhance the LBM region effectively. Approach. For this study, we originally developed a DE quantification (DEQ) method that uses a filter function to selectively provide modulated diffusion according to the original locations of objects in an image. The structural similarity index measure (SSIM) and Lie derivative image analysis L-value map were used to evaluate image quality and processing. Main results. We determined superellipse function with its order n = 4 as a better performing filter for the LBM region. DEQ was found to be more effective at contrasting LBM for various LBM CT images than PMD or its improved models when the filter was a positive exponential similar function. DEQ yields enhancement agreeing with the indications of positron emission tomography despite complex LBM comprising osteoblastic, osteoclastic, mixed tissues, and metal artifacts, which is innovative. Moreover, DEQ retained high quality of image (SSIM> 0.95), and achieved a low mean value of the L-value (Significance. Our method improved the visibility of mixed tissue lesions, which can assist computer visional framework and can help radiologists to produce accurate diagnose of LBM regions which are frequently overlooked in radiology findings because of the various degrees of visibility in CT images.
- 6 December 2024
- Published by IOP Publishing in Physics in Medicine & Biology
- Vol. 69 (24) , 245007
Open AccessDiffusion equation quantification: selective enhancement algorithm for bone metastasis lesions in CT images
Objective. Diffusion equation (DE) imaging processing is promising to enhance images showing lesions of bone metastasis (LBM). The Perona–Malik diffusion (PMD) model, which has been widely used and studied, is an anisotropic diffusion processing method to denoise or extract objects from an image effectively. However, the smoothing characteristics of PMD or its related method hinder extraction and enhancement of soft tissue regions of medical image such as computed tomography (CT), typically leaving an indistinct region with ambient tissues. Moreover, PMD expands the border region of the objects. A novel diffusion methodology must be used to enhance the LBM region effectively. Approach. For this study, we originally developed a DE quantification (DEQ) method that uses a filter function to selectively provide modulated diffusion according to the original locations of objects in an image. The structural similarity index measure (SSIM) and Lie derivative image analysis L-value map were used to evaluate image quality and processing. Main results. We determined superellipse function with its order n = 4 as a better performing filter for the LBM region. DEQ was found to be more effective at contrasting LBM for various LBM CT images than PMD or its improved models when the filter was a positive exponential similar function. DEQ yields enhancement agreeing with the indications of positron emission tomography despite complex LBM comprising osteoblastic, osteoclastic, mixed tissues, and metal artifacts, which is innovative. Moreover, DEQ retained high quality of image (SSIM> 0.95), and achieved a low mean value of the L-value (Significance. Our method improved the visibility of mixed tissue lesions, which can assist computer visional framework and can help radiologists to produce accurate diagnose of LBM regions which are frequently overlooked in radiology findings because of the various degrees of visibility in CT images.
- Open Access
- 11 December 2023
- Published by IOP Publishing in Physics in Medicine & Biology
- Vol. 68 (24) , 249501
Corrigendum: Extracting the gradient component of the gamma index using the Lie derivative method (2023 Phys. Med. Biol . 68 195028)
- 11 December 2023
- Published by IOP Publishing in Physics in Medicine & Biology
- Vol. 68 (24) , 249501
Open AccessCorrigendum: Extracting the gradient component of the gamma index using the Lie derivative method (2023 Phys. Med. Biol . 68 195028)
- Open Access
- 26 October 2023
- Published by Springer Nature in Research Square
Relative optimization force concept for gEUD biological optimization and novel a-value selection viewpoint
This document was written as the addendum of our published work, “Effective optimization strategy for large optimization volume object, remaining volume at risk (RVR): α-value selection and usage from generalized equivalent uniform dose (gEUD) curve deviation perspective” (10.1088/1361-6560/acb989). The gEUD is represented by a mathematical form of weighted general mean, and this is widely accepted as radiobiological dose in clinical practice. To make more smooth use and understanding of the gEUD optimization for the dose distribution, we focused on a-value roles for the optimization, and clarified the leverage effect of low and high a-value used optimization by means of relative optimization force (ROF), which is defined as a pseudo-force to explain the gEUD curve deformation.
- 26 October 2023
- Published by Springer Nature in Research Square
Open AccessRelative optimization force concept for gEUD biological optimization and novel a-value selection viewpoint
This document was written as the addendum of our published work, “Effective optimization strategy for large optimization volume object, remaining volume at risk (RVR): α-value selection and usage from generalized equivalent uniform dose (gEUD) curve deviation perspective” (10.1088/1361-6560/acb989). The gEUD is represented by a mathematical form of weighted general mean, and this is widely accepted as radiobiological dose in clinical practice. To make more smooth use and understanding of the gEUD optimization for the dose distribution, we focused on a-value roles for the optimization, and clarified the leverage effect of low and high a-value used optimization by means of relative optimization force (ROF), which is defined as a pseudo-force to explain the gEUD curve deformation.
- Open Access
- 5 October 2023
- Published by IOP Publishing in Physics in Medicine & Biology
- Vol. 68 (19) , 195028
Extracting the gradient component of the gamma index using the Lie derivative method
Objective . The gamma index ( γ ) has been extensively investigated in the medical physics and applied in clinical practice. However, γ has a significant limitation when used to evaluate the dose-gradient region, leading to inconveniences, particularly in stereotactic radiotherapy (SRT). This study proposes a novel evaluation method combined with γ to extract clinically problematic dose-gradient regions caused by irradiation including certain errors. Approach . A flow-vector field in the dose distribution is obtained when the dose is considered a scalar potential. Using the Lie derivative from differential geometry, we defined L , S , and U to evaluate the intensity, vorticity, and flow amount of deviation between two dose distributions, respectively. These metrics multiplied by γ ( γL , γS , γU ), along with the threshold value σ , were verified in the ideal SRT case and in a clinical case of irradiation near the brainstem region using radiochromic films. Moreover, Moran’s gradient index (MGI), Bakai’s χ factor, and the structural similarity index (SSIM) were investigated for comparisons. Main results . A high L- metric value mainly extracted high-dose-gradient induced deviations, which was supported by high S and U metrics observed as a robust deviation and an influence of the dose-gradient, respectively. The S- metric also denotes the measured similarity between the compared dose distributions. In the γ distribution, γL sensitively detected the dose-gradient region in the film measurement, despite the presence of noise. The threshold σ successfully extracted the gradient-error region where γ > 1 analysis underestimated, and σ = 0.1 (plan) and σ = 0.001 (film measurement) were obtained according to the compared resolutions. However, the MGI, χ, and SSIM failed to detect the clinically interested region. Significance . Although further studies are required to clarify the error details, this study demonstrated that the Lie derivative method provided a novel perspective for the identifying gradient-induced error regions and enabled enhanced and clinically significant evaluations of γ .
- 5 October 2023
- Published by IOP Publishing in Physics in Medicine & Biology
- Vol. 68 (19) , 195028
Open AccessExtracting the gradient component of the gamma index using the Lie derivative method
Objective . The gamma index ( γ ) has been extensively investigated in the medical physics and applied in clinical practice. However, γ has a significant limitation when used to evaluate the dose-gradient region, leading to inconveniences, particularly in stereotactic radiotherapy (SRT). This study proposes a novel evaluation method combined with γ to extract clinically problematic dose-gradient regions caused by irradiation including certain errors. Approach . A flow-vector field in the dose distribution is obtained when the dose is considered a scalar potential. Using the Lie derivative from differential geometry, we defined L , S , and U to evaluate the intensity, vorticity, and flow amount of deviation between two dose distributions, respectively. These metrics multiplied by γ ( γL , γS , γU ), along with the threshold value σ , were verified in the ideal SRT case and in a clinical case of irradiation near the brainstem region using radiochromic films. Moreover, Moran’s gradient index (MGI), Bakai’s χ factor, and the structural similarity index (SSIM) were investigated for comparisons. Main results . A high L- metric value mainly extracted high-dose-gradient induced deviations, which was supported by high S and U metrics observed as a robust deviation and an influence of the dose-gradient, respectively. The S- metric also denotes the measured similarity between the compared dose distributions. In the γ distribution, γL sensitively detected the dose-gradient region in the film measurement, despite the presence of noise. The threshold σ successfully extracted the gradient-error region where γ > 1 analysis underestimated, and σ = 0.1 (plan) and σ = 0.001 (film measurement) were obtained according to the compared resolutions. However, the MGI, χ, and SSIM failed to detect the clinically interested region. Significance . Although further studies are required to clarify the error details, this study demonstrated that the Lie derivative method provided a novel perspective for the identifying gradient-induced error regions and enabled enhanced and clinically significant evaluations of γ .
- 29 March 2023
- Published by Springer Nature in Japanese Journal of Radiology
- Vol. 41 (8) , 900-908
Deep learning-based detection of patients with bone metastasis from Japanese radiology reports
Purpose Deep learning (DL) is a state-of-the-art technique for developing artificial intelligence in various domains and it improves the performance of natural language processing (NLP). Therefore, we aimed to develop a DL-based NLP model that classifies the status of bone metastasis (BM) in radiology reports to detect patients with BM. Materials and Methods The DL-based NLP model was developed by training long short-term memory using 1,749 free-text radiology reports written in Japanese. We adopted five-fold cross-validation and used 200 reports for testing the five models. The accuracy, sensitivity, specificity, precision, and area under the receiver operating characteristics curve (AUROC) were used for the model evaluation. Results The developed model demonstrated classification performance with mean ± standard deviation of 0.912 ± 0.012, 0.924 ± 0.029, 0.901 ± 0.014, 0.898 ± 0.012, and 0.968 ± 0.004 for accuracy, sensitivity, specificity, precision, and AUROC, respectively. Conclusion The proposed DL-based NLP model may help in the early and efficient detection of patients with BM.
- 29 March 2023
- Published by Springer Nature in Japanese Journal of Radiology
- Vol. 41 (8) , 900-908
Deep learning-based detection of patients with bone metastasis from Japanese radiology reports
Purpose Deep learning (DL) is a state-of-the-art technique for developing artificial intelligence in various domains and it improves the performance of natural language processing (NLP). Therefore, we aimed to develop a DL-based NLP model that classifies the status of bone metastasis (BM) in radiology reports to detect patients with BM. Materials and Methods The DL-based NLP model was developed by training long short-term memory using 1,749 free-text radiology reports written in Japanese. We adopted five-fold cross-validation and used 200 reports for testing the five models. The accuracy, sensitivity, specificity, precision, and area under the receiver operating characteristics curve (AUROC) were used for the model evaluation. Results The developed model demonstrated classification performance with mean ± standard deviation of 0.912 ± 0.012, 0.924 ± 0.029, 0.901 ± 0.014, 0.898 ± 0.012, and 0.968 ± 0.004 for accuracy, sensitivity, specificity, precision, and AUROC, respectively. Conclusion The proposed DL-based NLP model may help in the early and efficient detection of patients with BM.
- 29 March 2023
- Published by Springer Nature in International Journal of Computer Assisted Radiology and Surgery
- Vol. 18 (10) , 1867-1874
Artificial intelligence-aided lytic spinal bone metastasis classification on CT scans
Purpose Spinal bone metastases directly affect quality of life, and patients with lytic-dominant lesions are at high risk for neurological symptoms and fractures. To detect and classify lytic spinal bone metastasis using routine computed tomography (CT) scans, we developed a deep learning (DL)-based computer-aided detection (CAD) system. Methods We retrospectively analyzed 2125 diagnostic and radiotherapeutic CT images of 79 patients. Images annotated as tumor (positive) or not (negative) were randomized into training (1782 images) and test (343 images) datasets. YOLOv5m architecture was used to detect vertebra on whole CT scans. InceptionV3 architecture with the transfer-learning technique was used to classify the presence/absence of lytic lesions on CT images showing the presence of vertebra. The DL models were evaluated via fivefold cross-validation. For vertebra detection, bounding box accuracy was estimated using intersection over union (IoU). We evaluated the area under the curve (AUC) of a receiver operating characteristic curve to classify lesions. Moreover, we determined the accuracy, precision, recall, and F1 score. We used the gradient-weighted class activation mapping (Grad-CAM) technique for visual interpretation. Results The computation time was 0.44 s per image. The average IoU value of the predicted vertebra was 0.923 ± 0.052 (0.684–1.000) for test datasets. In the binary classification task, the accuracy, precision, recall, F1-score, and AUC value for test datasets were 0.872, 0.948, 0.741, 0.832, and 0.941, respectively. Heat maps constructed using the Grad-CAM technique were consistent with the location of lytic lesions. Conclusion Our artificial intelligence-aided CAD system using two DL models could rapidly identify vertebra bone from whole CT images and detect lytic spinal bone metastasis, although further evaluation of diagnostic accuracy is required with a larger sample size.
- 29 March 2023
- Published by Springer Nature in International Journal of Computer Assisted Radiology and Surgery
- Vol. 18 (10) , 1867-1874
Artificial intelligence-aided lytic spinal bone metastasis classification on CT scans
Purpose Spinal bone metastases directly affect quality of life, and patients with lytic-dominant lesions are at high risk for neurological symptoms and fractures. To detect and classify lytic spinal bone metastasis using routine computed tomography (CT) scans, we developed a deep learning (DL)-based computer-aided detection (CAD) system. Methods We retrospectively analyzed 2125 diagnostic and radiotherapeutic CT images of 79 patients. Images annotated as tumor (positive) or not (negative) were randomized into training (1782 images) and test (343 images) datasets. YOLOv5m architecture was used to detect vertebra on whole CT scans. InceptionV3 architecture with the transfer-learning technique was used to classify the presence/absence of lytic lesions on CT images showing the presence of vertebra. The DL models were evaluated via fivefold cross-validation. For vertebra detection, bounding box accuracy was estimated using intersection over union (IoU). We evaluated the area under the curve (AUC) of a receiver operating characteristic curve to classify lesions. Moreover, we determined the accuracy, precision, recall, and F1 score. We used the gradient-weighted class activation mapping (Grad-CAM) technique for visual interpretation. Results The computation time was 0.44 s per image. The average IoU value of the predicted vertebra was 0.923 ± 0.052 (0.684–1.000) for test datasets. In the binary classification task, the accuracy, precision, recall, F1-score, and AUC value for test datasets were 0.872, 0.948, 0.741, 0.832, and 0.941, respectively. Heat maps constructed using the Grad-CAM technique were consistent with the location of lytic lesions. Conclusion Our artificial intelligence-aided CAD system using two DL models could rapidly identify vertebra bone from whole CT images and detect lytic spinal bone metastasis, although further evaluation of diagnostic accuracy is required with a larger sample size.
- 23 February 2023
- Published by IOP Publishing in Physics in Medicine & Biology
- Vol. 68 (5) , 055008
Effective optimization strategy for large optimization volume object, remaining volume at risk (RVR): α -value selection and usage from generalized equivalent uniform dose (gEUD) curve deviation perspective
Objective. A large optimization volume for intensity-modulated radiation therapy (IMRT), such as the remaining volume at risk (RVR), is traditionally unsuitable for dose-volume constraint control and requires planner-specific empirical considerations owing to the patient-specific shape. To enable less empirical optimization, the generalized equivalent uniform dose (gEUD) optimization is effective; however, the utilization of parameter a -values remains elusive. Our study clarifies the a -value characteristics for optimization and to enable effective a -value use. Approach. The gEUD can be obtained as a function of its a -value, which is the weighted generalized mean; its curve has a continuous, differentiable, and sigmoid shape, deforming in its optimization state with retained curve characteristics. Using differential geometry, the gEUD curve changes in optimization is considered a geodesic deviation intervened by the forces between deforming and retaining the curve. The curvature and gradient of the curve are radically related to optimization. The vertex point ( a = a k ) was set and the a -value roles were classified into the following three parts of the curve with respect to the a -value: (i) high gradient and middle curvature, (ii) middle gradient and high curvature, and (iii) low gradient and low curvature. Then, a strategy for multiple a -values was then identified using RVR optimization. Main results. Eleven head and neck patients who underwent static seven-field IMRT were used to verify the a -value characteristics and curvature effect for optimization. The lower a -value (i) ( a = 1–3) optimization was effective for the whole dose-volume range; in contrast, the effect of higher a -value (iii) ( a = 12–20) optimization addressed strongly the high-dose range of the dose volume. The middle a -value (ii) (around a = a k ) showed intermediate but effective high-to-low dose reduction. These a -value characteristics were observed as superimpositions in the optimization. Thus, multiple gEUD-based optimization was significantly superior to the exponential constraints normally applied to the RVR that surrounds the PTV, normal tissue objective (NTO), resulting in up to 25.9% and 8.1% improvement in dose-volume indices D2% and V10Gy, respectively. Significance. This study revealed an appropriate a -value for gEUD optimization, leading to favorable dose-volume optimization for the RVR region using fixed multiple a -value conditions, despite the very large and patient-specific shape of the region.
- 23 February 2023
- Published by IOP Publishing in Physics in Medicine & Biology
- Vol. 68 (5) , 055008
Effective optimization strategy for large optimization volume object, remaining volume at risk (RVR): α -value selection and usage from generalized equivalent uniform dose (gEUD) curve deviation perspective
Objective. A large optimization volume for intensity-modulated radiation therapy (IMRT), such as the remaining volume at risk (RVR), is traditionally unsuitable for dose-volume constraint control and requires planner-specific empirical considerations owing to the patient-specific shape. To enable less empirical optimization, the generalized equivalent uniform dose (gEUD) optimization is effective; however, the utilization of parameter a -values remains elusive. Our study clarifies the a -value characteristics for optimization and to enable effective a -value use. Approach. The gEUD can be obtained as a function of its a -value, which is the weighted generalized mean; its curve has a continuous, differentiable, and sigmoid shape, deforming in its optimization state with retained curve characteristics. Using differential geometry, the gEUD curve changes in optimization is considered a geodesic deviation intervened by the forces between deforming and retaining the curve. The curvature and gradient of the curve are radically related to optimization. The vertex point ( a = a k ) was set and the a -value roles were classified into the following three parts of the curve with respect to the a -value: (i) high gradient and middle curvature, (ii) middle gradient and high curvature, and (iii) low gradient and low curvature. Then, a strategy for multiple a -values was then identified using RVR optimization. Main results. Eleven head and neck patients who underwent static seven-field IMRT were used to verify the a -value characteristics and curvature effect for optimization. The lower a -value (i) ( a = 1–3) optimization was effective for the whole dose-volume range; in contrast, the effect of higher a -value (iii) ( a = 12–20) optimization addressed strongly the high-dose range of the dose volume. The middle a -value (ii) (around a = a k ) showed intermediate but effective high-to-low dose reduction. These a -value characteristics were observed as superimpositions in the optimization. Thus, multiple gEUD-based optimization was significantly superior to the exponential constraints normally applied to the RVR that surrounds the PTV, normal tissue objective (NTO), resulting in up to 25.9% and 8.1% improvement in dose-volume indices D2% and V10Gy, respectively. Significance. This study revealed an appropriate a -value for gEUD optimization, leading to favorable dose-volume optimization for the RVR region using fixed multiple a -value conditions, despite the very large and patient-specific shape of the region.