Advantages of Persistent Cohomology in Estimating Animal Location From Grid Cell Population Activity.

Many cognitive functions are represented as cell assemblies. In the case of spatial navigation, the population activity of place cells in the hippocampus and grid cells in the entorhinal cortex represents self-location in the environment. The brain cannot directly observe self-location information in the environment. Instead, it relies on sensory information and memory to estimate self-location. Therefore, estimating low-dimensional dynamics, such as the movement trajectory of an animal exploring its environment, from only the high-dimensional neural activity is important in deciphering the information represented in the brain. Most previous studies have estimated the low-dimensional dynamics (i.e., latent variables) behind neural activity by unsupervised learning with Bayesian population decoding using artificial neural networks or gaussian processes. Recently, persistent cohomology has been used to estimate latent variables from the phase information (i.e., circular coordinates) of manifolds created by neural activity. However, the advantages of persistent cohomology over Bayesian population decoding are not well understood. We compared persistent cohomology and Bayesian population decoding in estimating the animal location from simulated and actual grid cell population activity. We found that persistent cohomology can estimate the animal location with fewer neurons than Bayesian population decoding and robustly estimate the animal location from actual noisy data.

Dosiomics for intensity-modulated radiotherapy in patients with prostate cancer: survival analysis stratified by baseline prostate-specific antigen and Gleason grade group in a 2-institutional retrospective study.

OBJECTIVE: This study evaluated the prognostic impact of the quality of dose distribution using dosiomics in patients with prostate cancer, stratified by pretreatment prostate-specific antigen (PSA) levels and Gleason grade (GG) group. METHODS: A total of 721 patients (Japanese Foundation for Cancer Research [JFCR] cohort: N = 489 and Tokyo Radiation Oncology Clinic [TROC] cohort: N = 232) with localized prostate cancer treated by intensity-modulated radiation therapy were enrolled. Two predictive dosiomic features for biochemical recurrence (BCR) were selected and patients were divided into certain groups stratified by pretreatment PSA levels and GG. Freedom from biochemical failure (FFBF) was estimated using the Kaplan-Meier method based on each dosiomic feature and univariate discrimination was evaluated using the log-rank test. As an exploratory analysis, a dosiomics hazard (DH) score was developed and its prognostic power for BCR was examined. RESULTS: The dosiomic feature extracted from planning target volume (PTV) significantly distinguished the high- and low-risk groups in patients with PSA levels >10 ng/mL (7-year FFBF: 86.7% vs 76.1%, P < .01), GG 4 (92.2% vs 76.9%, P < .01), and GG 5 (83.1% vs 77.8%, P = .04). The DH score showed significant association with BCR (hazard score: 2.04; 95% confidence interval: 1.38-3.01; P < .001). CONCLUSION: The quality of planned dose distribution on PTV may affect the prognosis of patients with poor prognostic factors, such as PSA levels >10 ng/mL and higher GGs. ADVANCES IN KNOWLEDGE: The effects of planned dose distribution on prognosis differ depending on the patient's clinical background.

A simulation study on the radiation-induced immune response of tumors after single fraction high-dose irradiation.

PURPOSE: We investigated radiation-induced antitumor immunity and its suppression by hypoxia-inducible factor (HIF-1α) for radiosurgery (SRS) using an improved cellular automata (CA) model. METHOD: A two-dimensional Cellular Automata (CA) model was employed to simulate the impact of radiation on cancer cell death and subsequent immune responses. Cancer cells died from direct cell death from radiation and indirect cell death due to radiation-induced vascular damage. The model also incorporated radiation-induced immunity and immuno-suppression. It was incorporated into the model assuming that the death of cancer cells generates effector cells, forming complexes with cancer cells, and high radiation doses lead to vascular damage, inducing tumor hypoxia and increasing HIF-1α expression. The model was validated and subjected to sensitivity analysis by evaluating tumor volume changes post-irradiation and exploring the effects and sensitivity of radiation-induced immune responses. RESULTS: The ratios of the tumor volume at 360 days post-irradiation and the SRS day (rTV) decreased with a higher PME, a higher Pcomp, and a lower ThHIF. The rTVs were 4.6 and 2.0 for PME = 0.1 and 0.9, 12.0 and 2.2 for Pcomp = 0.1 and 0.9, and 1.5 and 15.3 for ThHIF = 0.1 and 10.0, respectively. CONCLUSIONS: By modeling the activation and deactivation of the effectors, the improved CA model showed that the radiation-induced immunogenic cell death in the tumor caused a decrease in the post-irradiation volume by a factor of four for the therapeutic doses relative to non-immune reaction cases. Furthermore, the suppressive effects of HIF-1α induced by hypoxia decreased radiation-induced immune effects by more than 50.

Radiomics and dosiomics for predicting complete response to definitive chemoradiotherapy patients with oesophageal squamous cell cancer using the hybrid institution model.

OBJECTIVES: To develop a multi-institutional prediction model to estimate the local response to oesophageal squamous cell carcinoma (ESCC) treated with definitive radiotherapy based on radiomics and dosiomics features. METHODS: The local responses were categorised into two groups (incomplete and complete). An external validation model and a hybrid model that the patients from two institutions were mixed randomly were proposed. The ESCC patients at stages I-IV who underwent chemoradiotherapy from 2012 to 2017 and had follow-up duration of more than 5 years were included. The patients who received palliative or pre-operable radiotherapy and had no FDG PET images were excluded. The segmentations included the GTV, CTV, and PTV which are used in treatment planning. In addition, shrinkage, expansion, and shell regions were created. Radiomic and dosiomic features were extracted from CT, FDG PET images, and dose distribution. Machine learning-based prediction models were developed using decision tree, support vector machine, k-nearest neighbour (kNN) algorithm, and neural network (NN) classifiers. RESULTS: A total of 116 and 26 patients enrolled at Centre 1 and Centre 2, respectively. The external validation model exhibited the highest accuracy with 65.4% for CT-based radiomics, 77.9% for PET-based radiomics, and 72.1% for dosiomics based on the NN classifiers. The hybrid model exhibited the highest accuracy of 84.4% for CT-based radiomics based on the kNN classifier, 86.0% for PET-based radiomics, and 79.0% for dosiomics based on the NN classifiers. CONCLUSION: The proposed hybrid model exhibited promising predictive performance for the local response to definitive radiotherapy in ESCC patients. CLINICAL RELEVANCE STATEMENT: The prediction of the complete response for oesophageal cancer patients may contribute to improving overall survival. The hybrid model has the potential to improve prediction performance than the external validation model that was conventionally proposed. KEY POINTS: • Radiomics and dosiomics used to predict response in patients with oesophageal cancer receiving definitive radiotherapy. • Hybrid model with neural network classifier of PET-based radiomics improved prediction accuracy by 8.1%. • The hybrid model has the potential to improve prediction performance.

Artificial intelligence-based diagnosis of the depth of laryngopharyngeal cancer.

OBJECTIVE: Transoral surgery (TOS) is a widely used treatment for laryngopharyngeal cancer. There are some difficult cases of setting the extent of resection in TOS, particularly in setting the vertical margins. However, positive vertical margins require additional treatment. Further, excessive resection should be avoided as it increases the risk of bleeding as a postoperative complication and may lead to decreased quality of life, such as dysphagia. Considering these issues, determining the extent of resection in TOS is an important consideration. In this study, we investigated the possibility of accurately diagnosing the depth of laryngopharyngeal cancer using radiomics, an image analysis method based on artificial intelligence (AI). METHODS: We included esophagogastroduodenoscopic images of 95 lesions that were pathologically diagnosed as squamous cell carcinoma (SCC) and treated with transoral surgery at our institution between August 2009 and April 2020. Of the 95 lesions, 54 were SCC in situ, and 41 were SCC. Radiomics analysis was performed on 95 upper gastrointestinal endoscopic NBI images of these lesions to evaluate their diagnostic performance for the presence of subepithelial invasion. The lesions in the endoscopic images were manually delineated, and the accuracy, sensitivity, specificity, and area under the curve (AUC) were evaluated from the features obtained using least absolute shrinkage and selection operator analysis. In addition, the results were compared with the depth predictions made by skilled endoscopists. RESULTS: In the Radiomics study, the average cross-validation was 0.833. The mean AUC for cross-validation calculated from the receiver operating characteristic curve was 0.868. These results were equivalent to those of the diagnosis made by a skilled endoscopist. CONCLUSION: The diagnosis of laryngopharyngeal cancer depth using radiomics analysis has potential clinical applications. We plan to use it in actual surgery in the future and prospectively study whether it can be used for diagnosis.

Unbiased evaluation of predicted gamma passing rate by an event-mixing technique.

BACKGROUND: Predicting models of the gamma passing rate (GPR) have been studied to substitute the measurement-based gamma analysis. Since these studies used data from different radiotherapy systems comprising TPS, linear accelerator, and detector array, it has been difficult to compare the performances of the predicting models among institutions with different radiotherapy systems. PURPOSE: We aimed to develop unbiased scoring methods to evaluate the performance of the models predicting the GPR, by introducing both best and worst limits for the performance of the GPR prediction. METHODS: Two hundred head-and-neck VMAT plans were used to develop a framework. The GPRs were measured using the ArcCHECK device. The predicted GPR [p] was generated using a deep learning-based model [pDL ]. The predicting model was evaluated using four metrics: standard deviation (SD) [σ], Pearson's correlation coefficient (CC) [r], mean squared error (MSE) [s], and mean absolute error (MAE) [a]. The best limit [ σ m ${\sigma _m}$ , r m ${r_m}$ , s m ${s_m}$ , and a m ${a_m}$ ] was estimated by measuring the SD of measured GPR [m] by shifting the device along the longitudinal direction to measure different sampling points. Mimicked best and worst p's [pbest and pworst ] were generated from pDL . The worst limit was defined such that m and p have no correlation [CC ∼ 0]. The worst limit [σMix , rMix , sMix , and aMix ] was generated using the event-mixing (EM) technique originally introduced in high-energy physics experiments. The range of σ, r, s, and a was defined to be [ σ m , σ Mix ] $[ {{\sigma _m},{\sigma _{{\mathrm{Mix}}}}} ]$ , [ 0 , r m ] $[ {0,{r_m}} ]$ , [ s m , s Mix ] $[ {{s_m},{s_{{\mathrm{Mix}}}}} ]$ , and [ a m , a Mix ] $[ {{a_m},{a_{{\mathrm{Mix}}}}} ]$ . The achievement score (AS) independently based on σ, r, s, and a were calculated for pDL , pbest and pworst . The probability that p fails the gamma analysis (alert frequency; AF) was estimated as a function of σ d ${\sigma _d}$ values within the [ σ m ${\sigma _m}$ , σMix ] range for the 3%/2 mm data with a 95% criterion. RESULTS: SDs of the best limit were well reproduced by σ m = 0.531 100 - m ${\sigma _m} = \;0.531\sqrt {100 - m} $ . The EM technique successfully generated the ( m , p ) $( {m,p} )$ pairs with no correlation. The AS using four metrics showed good agreement. This agreement indicates successful definitions of both best and worst limits, consistent definitions of the AS, and successful generations of mixed events. The AF for the DL-based model with the 3%/2 mm tolerance was 31.5% and 63.0% with CL's 99% and 99.9%, respectively. CONCLUSION: We developed the AS to evaluate the predicting model of the GPR in an unbiased manner by excluding the effects of the precision of the radiotherapy system and the spreading of the GPR. The best and worst limits of the GPR prediction were successfully generated using the measured precision of the GPR and the EM technique, respectively. The AS and σ p ${\sigma _p}$ are expected to enable objective evaluation of the predicting model and setting exact achievement goal of precision for the predicted GPR.

CT number calibration audit in photon radiation therapy.

BACKGROUND: Inadequate computed tomography (CT) number calibration curves affect dose calculation accuracy. Although CT number calibration curves registered in treatment planning systems (TPSs) should be consistent with human tissues, it is unclear whether adequate CT number calibration is performed because CT number calibration curves have not been assessed for various types of CT number calibration phantoms and TPSs. PURPOSE: The purpose of this study was to investigate CT number calibration curves for mass density (ρ) and relative electron density (ρe ). METHODS: A CT number calibration audit phantom was sent to 24 Japanese photon therapy institutes from the evaluating institute and scanned using their individual clinical CT scan protocols. The CT images of the audit phantom and institute-specific CT number calibration curves were submitted to the evaluating institute for analyzing the calibration curves registered in the TPSs at the participating institutes. The institute-specific CT number calibration curves were created using commercial phantom (Gammex, Gammex Inc., Middleton, WI, USA) or CIRS phantom (Computerized Imaging Reference Systems, Inc., Norfolk, VA, USA)). At the evaluating institute, theoretical CT number calibration curves were created using a stoichiometric CT number calibration method based on the CT image, and the institute-specific CT number calibration curves were compared with the theoretical calibration curve. Differences in ρ and ρe over the multiple points on the curve (Δρm and Δρe,m , respectively) were calculated for each CT number, categorized for each phantom vendor and TPS, and evaluated for three tissue types: lung, soft tissues, and bones. In particular, the CT-ρ calibration curves for Tomotherapy TPSs (ACCURAY, Sunnyvale, CA, USA) were categorized separately from the Gammex CT-ρ calibration curves because the available tissue-equivalent materials (TEMs) were limited by the manufacturer recommendations. In addition, the differences in ρ and ρe for the specific TEMs (ΔρTEM and Δρe,TEM , respectively) were calculated by subtracting the ρ or ρe of the TEMs from the theoretical CT-ρ or CT-ρe calibration curve. RESULTS: The mean ± standard deviation (SD) of Δρm and Δρe,m for the Gammex phantom were -1.1 ± 1.2 g/cm3 and -0.2 ± 1.1, -0.3 ± 0.9 g/cm3 and 0.8 ± 1.3, and -0.9 ± 1.3 g/cm3 and 1.0 ± 1.5 for lung, soft tissues, and bones, respectively. The mean ± SD of Δρm and Δρe,m for the CIRS phantom were 0.3 ± 0.8 g/cm3 and 0.9 ± 0.9, 0.6 ± 0.6 g/cm3 and 1.4 ± 0.8, and 0.2 ± 0.5 g/cm3 and 1.6 ± 0.5 for lung, soft tissues, and bones, respectively. The mean ± SD of Δρm for Tomotherapy TPSs was 2.1 ± 1.4 g/cm3 for soft tissues, which is larger than those for other TPSs. The mean ± SD of Δρe,TEM for the Gammex brain phantom (BRN-SR2) was -1.8 ± 0.4, implying that the tissue equivalency of the BRN-SR2 plug was slightly inferior to that of other plugs. CONCLUSIONS: Latent deviations between human tissues and TEMs were found by comparing the CT number calibration curves of the various institutes.

Biological dosimetric impact of dose-delivery time for hypoxic tumour with modified microdosimetric kinetic model.

Background: An improved microdosimetric kinetic model (MKM) can address radiobiological effects with prolonged delivery times. However, these do not consider the effects of oxygen. The current study aimed to evaluate the biological dosimetric effects associated with the dose delivery time in hypoxic tumours with improved MKM for photon radiation therapy. Materials and methods: Cell survival was measured under anoxic, hypoxic, and oxic conditions using the Monte Carlo code PHITS. The effect of the dose rate of 0.5-24 Gy/min for the biological dose (Dbio) was estimated using the microdosimetric kinetic model. The dose per fraction and pressure of O2 (pO2) in the tumour varied from 2 to 20 Gy and from 0.01 to 5.0% pO2, respectively. Results: The ratio of the Dbio at 1.0-24 Gy/min to that at 0.5 Gy/min (RDR) was higher at higher doses. The maximum RDR was 1.09 at 1.0 Gy/min, 1.12 at 12 Gy/min, and 1.13 at 24 Gy/min. The ratio of the Dbio at 0.01-2.0% of pO2 to that at 5.0% of pO2 (Roxy) was within 0.1 for 2-20 Gy of physical dose. The maximum Roxy was 0.42 at 0.01% pO2, 0.76 at 0.4% pO2, 0.89 at 1% pO2, and 0.96 at 2% pO2. Conclusion: Our proposed model can estimate the cell killing and biological dose under hypoxia in a clinical and realistic patient. A shorter dose-delivery time with a higher oxygen distribution increased the radiobiological effect. It was more effective at higher doses per fraction than at lower doses.

Novel simulation for dosimetry impact of diaphragm respiratory motion in four-dimensional volumetric modulated arc therapy for esophageal cancer.

BACKGROUND AND PURPOSE: The diaphragm respiratory motion (RM) could impact the target dose robustness in the lower esophageal cancer (EC). We aimed to develop a framework evaluating the impact of different RM patterns quantitatively in one patient, by creating virtual four-dimensional computed-tomography (v4DCT) images, which could lead to tailored treatment for the breathing pattern. We validated virtual 4D radiotherapy (v4DRT) along with exploring the acceptability of free-breathing volumetric modulated arc therapy (FB-VMAT). METHODS AND MATERIALS: We assessed 10 patients with superficial EC through their real 4DCT (r4DCT) scans. v4DCT images were derived from the end-inhalation computed tomography (CT) image (reference CT) and the v4DRT dose was accumulated dose over all phases. r4DRT diaphragm shifts were applied with magnitudes derived from r4DCT scans; clinical target volume (CTV) dose of v4DRT was compared with that of r4DRT to validate v4DRT. CTV dosage modifications and planning organ at risk volume (PRV) margins of the spinal cord were examined with the diaphragm movement. The percentage dose differences (ΔDx) were determined between the v4DRT and the dose calculated on the reference CT image. RESULTS: The CTV ΔDx between the r4DRT and v4DRT were within 1% in cases with RM ≦ 15 mm. The average ΔD100% and ΔDmean of the CTV ranging from 5 to 15 mm of diaphragm motion was 0.3% to 1.7% and 0.1% to 0.4%, respectively. All CTV index changes were within 3% and ΔD1cc and ΔD2cc of Cord PRV were within 1%. CONCLUSION: We postulate a novel method for evaluating the CTV robustness, comparable to the conventional r4DCT method under the diaphragm RM ≦ 15 mm permitting an impact of within 3% in FB-VMAT for EC on the CTV dose distribution.

Dose uncertainty due to energy dependence in dual-energy computed tomography.

Purpose: To evaluate the absolute dose uncertainty at 2 different energies and for the large and small bowtie filters in dual-energy computed tomography (DECT). Material and methods: Measurements were performed using DECT at 80 kV and 140 kilovoltage peak (kVp), and single-energy computed tomography (CT) at 120 kV. The absolute dose was calculated from the mass-energy absorption obtained from the half-value layer (HVL) of aluminium. Results: The difference in the water-to-air ratio of the mean mass energy-absorption coefficients at 80 kV and 140 kV was 2.0% for the small bow-tie filter and 3.0% for the large bow-tie filter. At lower tube voltages, the difference in the absorbed dose with the large and small bow-tie filters was larger. Conclusions: The absolute dose uncertainty due to energy dependence was 3.0%, which could be reduced with single-energy beams at 120 kV or by using the average effective energy measurement with dual-energy beams.

Prognostic Value of Radiomics Analysis of Skeletal Muscle After Radical Irradiation of Esophageal Cancer.

BACKGROUND/AIM: Sarcopenia is an independent survival predictor in several tumor types. Computed tomography (CT) is the standard measurement for body composition assessment. Radiomics analysis of CT images allows for the precise evaluation of skeletal muscles. This study aimed to construct a prognostic survival model for patients with esophageal cancer who underwent radical irradiation using skeletal muscle radiomics. PATIENTS AND METHODS: We retrospectively identified patients with esophageal cancer who underwent radical irradiation at our institution between April 2008 and December 2017. Skeletal muscle radiomics were extracted from an axial pretreatment CT at the third lumbar vertebral level. The prediction model was constructed using machine learning coupled with the least absolute shrinkage and selection operator (LASSO). The predictive nomogram model comprised clinical factors with radiomic features. Three prediction models were created: clinical, radiomics, and combined. RESULTS: Ninety-eight patients with 98 esophageal cancers were enrolled in this study. The median observation period was 57.5 months (range=1-98 months). Thirty-five radiomics features were selected by LASSO analysis, and a prediction model was constructed using training and validation data. The average of the accuracy, specificity, sensitivity, and area under the concentration-time curve for predicting survival in esophageal cancer in the combined model were 75%, 92%, and 0.86, respectively. The C-indices of the clinical, radiomics, and combined models were 0.76, 0.80, and 0.88, respectively. CONCLUSION: A prediction model with skeletal muscle radiomics and clinical data might help determine survival outcomes in patients with esophageal cancer treated with radical radiotherapy.

Image masking using convolutional networks improves performance classification of radiation pneumonitis for non-small cell lung cancer.

Radiation pneumonitis (RP) is a serious side effect of radiotherapy in patients with locally advanced non-small-cell lung cancer (NSCLC). Image cropping reduces training noise and may improve classification accuracy. This study proposes a prediction model for RP grade ≥ 2 using a convolutional neural network (CNN) model with image cropping. The 3D computed tomography (CT) images cropped in the whole-body, normal lung (nLung), and nLung regions overlapping the region over 20 Gy (nLung∩20 Gy) used in treatment planning were used as the input data. The output classifies patients as RP grade < 2 or RP grade ≥ 2. The sensitivity, specificity, accuracy, and area under the curve (AUC) were evaluated using the receiver operating characteristic curve (ROC). The accuracy, specificity, sensitivity, and AUC were 53.9%, 80.0%, 25.5%, and 0.58, respectively, for the whole-body method, and 60.0%, 81.7%, 36.4%, and 0.64, respectively, for the nLung method. For the nLung∩20 Gy method, the accuracy, specificity, sensitivity, and AUC improved to 75.7%, 80.0%, 70.9%, and 0.84, respectively. The CNN model, in which the input image is segmented in the normal lung considering the dose distribution, can help predict an RP grade ≥ 2 for NSCLC patients after definitive radiotherapy.

Physical and biological dosimetric margin according to prescription method for stereotactic body radiation therapy.

This study aimed to expand the biological conversion factor (BCF) model, which converts the physical dosimetric margin (PDM) to the biological dosimetric margin (BDM) for point prescription with 3-dimensional conformal radiation therapy (3DCRT) and the marginal prescription method with volumetric-modulated arc radiotherapy (VMAT). The VMAT of the marginal prescription and the 3DCRT of the point prescription with lung stereotactic body radiation therapy (SBRT) by using RayStation were planned. The biological equivalent dose (BED) for a dose per fraction (DPF) of 3-20 Gy was calculated from these plans. The dose was perturbed with the calculation using a 1-mm step isocenter shift. The dose covering 95% of the target was greater than or equal to 90% of the prescribed physical dose, and the BED were defined as the PDM and BDM, respectively. The BCF was created as a function of the DPF. The PDM and BDM for all DPFs were larger with the point prescription method than with the marginal prescription method. The marginal prescription method with a 60% isodose line had a larger PDM and BDM. The BCF with the point prescription was smaller than that with the marginal prescription in the left-right (LR), anterior-posterior (AP) and cranio-caudal (CC) directions. In the marginal prescription method, the 60% isodose line had a higher BCF. In conclusion, the improved BCF method could be converted to BDM for point prescription with 3DCRT and marginal prescription method with VMAT, which is required for stereotactic radiation therapy in radiobiology-based treatment planning.

A synthesized gamma distribution-based patient-specific VMAT QA using a generative adversarial network.

BACKGROUND: Artificial intelligence (AI)-based gamma passing rate (GPR) prediction has been proposed as a time-efficient virtual patient-specific QA method for the delivery of volumetric modulation arc therapy (VMAT). However, there is a limitation that the GPR value loses the locational information of dose accuracy. PURPOSE: The objective was to predict the failing points in the gamma distribution and the GPR using a synthesized gamma distribution of VMAT QA with a deep convolutional generative adversarial network (GAN). METHODS: The fluence maps of 270 VMAT beams for prostate cancer were measured using an electronic portal imaging device and analyzed using gamma evaluation with 3%/2-mm, 2%/1-mm, 1%/1-mm, and 1%/0.5-mm tolerances. The 270 gamma distributions were divided into two datasets: 240 training datasets for creating a model and 30 test datasets for evaluation. The image prediction network for the fluence maps calculated by the treatment planning system (TPS) to the gamma distributions was created using a GAN. The sensitivity, specificity, and accuracy of detecting failing points were evaluated using measured and synthesized gamma distributions. In addition, the difference between measured GPR (mGPR) and predicted GPR (pGPR) values calculated from the synthesized gamma distributions was evaluated. RESULTS: The root mean squared errors between mGPR and pGPR were 1.0%, 2.1%, 3.5%, and 3.6% for the 3%/2-mm, 2%/1-mm, 1%/1-mm, and 1%/0.5-mm tolerances, respectively. The accuracies for detecting failing points were 98.9%, 96.9%, 94.7%, and 93.7% for 3%/2-mm, 2%/1-mm, 1%/1-mm, and 1%/0.5-mm tolerances, respectively. The sensitivity and specificity were the highest for 1%/0.5-mm and 3%/2-mm tolerances, which were 82.7% and 99.6%, respectively. CONCLUSIONS: We developed a novel system using a GAN to generate a synthesized gamma distribution-based patient-specific VMAT QA. The system is promising from the point of view of quality assurance in radiotherapy because it shows high performance and can detect failing points.

MRI image synthesis for fluid-attenuated inversion recovery and diffusion-weighted images with deep learning.

This study aims to synthesize fluid-attenuated inversion recovery (FLAIR) and diffusion-weighted images (DWI) with a deep conditional adversarial network from T1- and T2-weighted magnetic resonance imaging (MRI) images. A total of 1980 images of 102 patients were split into two datasets: 1470 (68 patients) in a training set and 510 (34 patients) in a test set. The prediction framework was based on a convolutional neural network with a generator and discriminator. T1-weighted, T2-weighted, and composite images were used as inputs. The digital imaging and communications in medicine (DICOM) images were converted to 8-bit red-green-blue images. The red and blue channels of the composite images were assigned to 8-bit grayscale pixel values in T1-weighted images, and the green channel was assigned to those in T2-weighted images. The prediction FLAIR and DWI images were of the same objects as the inputs. For the results, the prediction model with composite MRI input images in the DWI image showed the smallest relative mean absolute error (rMAE) and largest mutual information (MI), and that in the FLAIR image showed the largest relative mean-square error (rMSE), relative root-mean-square error (rRMSE), and peak signal-to-noise ratio (PSNR). For the FLAIR image, the prediction model with the T2-weighted MRI input images generated more accurate synthesis results than that with the T1-weighted inputs. The proposed image synthesis framework can improve the versatility and quality of multi-contrast MRI without extra scans. The composite input MRI image contributes to synthesizing the multi-contrast MRI image efficiently.

Influence of different air CT numbers for IVDT on the dose distribution in TomoTherapy MVCT.

This study aims to evaluate the effect of different air computed tomography (CT) numbers of the image value density table (IVDT) on the retrospective dose calculation of head-and-neck (HN) radiotherapy using TomoTherapy megavoltage CT (MVCT) images. The CT numbers of the inside and outside air and each tissue-equivalent plug of the "Cheese" phantom were obtained from TomoTherapy MVCT. Two IVDTs with different air CT numbers were created and applied to MVCT images of the HN anthropomorphic phantom and recalculated by Planned Adaptive to verify dose distribution. We defined the recalculation dose with MVCT images using both inside and outside air of the IVDT as IVDT MVCT inair ${\mathrm{IVDT}}_{\mathrm{MVCT}}^{\mathrm{inair}}$ and IVDT MVCT outair ${\mathrm{IVDT}}_{\mathrm{MVCT}}^{\mathrm{outair}}$ , respectively. Treatment planning doses calculated on kVCT images were compared with those calculated on MVCT images using two different IVDT tables, namely, IVDT MVCT inair ${\mathrm{IVDT}}_{\mathrm{MVCT}}^{\mathrm{inair}}$ and IVDT MVCT outair ${\mathrm{IVDT}}_{\mathrm{MVCT}}^{\mathrm{outair}}$ . The difference between average MVCT numbers ±1 standard deviation on inside and outside air of the calibration phantom was 65 ± 36 HU. This difference in MVCT number of air exceeded the recommendation lung tolerance for dose calculation error of 2%. The dose differences between the planning target volume (PTV): D98% , D50% , D2% and the organ at risk (OAR): Dmax , Dmean recalculated by IVDT MVCT inair ${\mathrm{IVDT}}_{\mathrm{MVCT}}^{\mathrm{inair}}$ and IVDT MVCT outair ${\mathrm{IVDT}}_{\mathrm{MVCT}}^{\mathrm{outair}}$ using MVCT images were a maximum of 0.7% and 1.2%, respectively. Recalculated doses to the PTV and OAR with MVCT showed that IVDT MVCT outair ${\mathrm{IVDT}}_{\mathrm{MVCT}}^{\mathrm{outair}}$ was 0.5%-0.7% closer to the kVCT treatment planning dose than IVDT MVCT inair ${\mathrm{IVDT}}_{\mathrm{MVCT}}^{\mathrm{inair}}$ . This study showed that IVDT MVCT outair ${\mathrm{IVDT}}_{\mathrm{MVCT}}^{\mathrm{outair}}$ was more accurate than IVDT MVCT inair ${\mathrm{IVDT}}_{\mathrm{MVCT}}^{\mathrm{inair}}$ in recalculating the dose HN cases of MVCT using TomoTherapy.

Image synthesis of effective atomic number images using a deep convolutional neural network-based generative adversarial network.

Background: The effective atomic numbers obtained from dual-energy computed tomography (DECT) can aid in characterization of materials. In this study, an effective atomic number image reconstructed from a DECT image was synthesized using an equivalent single-energy CT image with a deep convolutional neural network (CNN)-based generative adversarial network (GAN). Materials and methods: The image synthesis framework to obtain the effective atomic number images from a single-energy CT image at 120 kVp using a CNN-based GAN was developed. The evaluation metrics were the mean absolute error (MAE), relative root mean square error (RMSE), relative mean square error (MSE), structural similarity index (SSIM), peak signal-to-noise ratio (PSNR), and mutual information (MI). Results: The difference between the reference and synthetic effective atomic numbers was within 9.7% in all regions of interest. The averages of MAE, RMSE, MSE, SSIM, PSNR, and MI of the reference and synthesized images in the test data were 0.09, 0.045, 0.0, 0.89, 54.97, and 1.03, respectively. Conclusions: In this study, an image synthesis framework using single-energy CT images was constructed to obtain atomic number images scanned by DECT. This image synthesis framework can aid in material decomposition without extra scans in DECT.

Improved biological dosimetric margin model for different PTV margins with stereotactic body radiation therapy in homogeneous and nonhomogeneous tumor regions.

Background: The purpose of this study was to improve the biological dosimetric margin (BDM) corresponding to different planning target volume (PTV) margins in homogeneous and nonhomogeneous tumor regions using an improved biological conversion factor (BCF) model for stereotactic body radiation therapy (SBRT). Materials and methods: The PTV margin was 5-20 mm from the clinical target volume. The biologically equivalent dose (BED) was calculated using the linear-quadratic model. The biological parameters were α/ß = 10 Gy, and the dose per fraction (DPF) was d = 3-20 Gy/fr. The isocenter was offset at intervals of 1 mm; 95% of the clinical target volume covered more than 90% of the prescribed physical dose, and BED was defined as biological and physical DMs. The BCF formula was defined as a function of the DPF. Results: The difference in the BCF caused by the DPF was within 0.05 for the homogeneous and nonhomogeneous phantoms. In the virtual nonhomogeneous phantom, the data with a PTV margin of 10-20 mm were not significantly different; thus, these were combined to fit the BCF. In the virtual homogeneous phantom, the BCF was fitted to each PTV margin. Conclusions: The current study improved a scheme to estimate the BDM considering the size of the PTV margin and homogeneous and nonhomogeneous regions. This technique is expected to enable BED-based treatment planning using treatment systems based on physical doses for SBRT.

Radiomic Analysis for Pretreatment Prediction of Recurrence Post-Radiotherapy in Cervical Squamous Cell Carcinoma Cancer.

BACKGROUND: The current study aims to predict the recurrence of cervical cancer patients treated with radiotherapy from radiomics features on pretreatment T1- and T2-weighted MR images. METHODS: A total of 89 patients were split into model training (63 patients) and model testing (26 patients). The predictors of recurrence were selected using the least absolute shrinkage and selection operator (LASSO) regression. The machine learning used neural network classifiers. RESULTS: Using LASSO analysis of radiomics, we found 25 features from the T1-weighted and 4 features from T2-weighted MR images, respectively. The accuracy was highest with the combination of T1- and T2-weighted MR images. The model performances with T1- or T2-weighted MR images were 86.4% or 89.4% accuracy, 74.9% or 38.1% sensitivity, 81.8% or 72.2% specificity, and 0.89 or 0.69 of the area under the curve (AUC). The model performance with the combination of T1- and T2-weighted MR images was 93.1% accuracy, 81.6% sensitivity, 88.7% specificity, and 0.94 of AUC. CONCLUSIONS: The radiomics analysis with T1- and T2-weighted MR images could highly predict the recurrence of cervix cancer after radiotherapy. The variation of the distribution and the difference in the pixel number at the peripheral and the center were important predictors.

Predictive gamma passing rate of 3D detector array-based volumetric modulated arc therapy quality assurance for prostate cancer via deep learning.

To predict the gamma passing rate (GPR) of the three-dimensional (3D) detector array-based volumetric modulated arc therapy (VMAT) quality assurance (QA) for prostate cancer using a convolutional neural network (CNN) with the 3D dose distribution. One hundred thirty-five VMAT plans for prostate cancer were selected: 110 plans were used for training and validation, and 25 plans were used for testing. Verification plans were measured using a helical 3D diode array (ArcCHECK). The dose distribution on the detector element plane of these verification plans was used as input data for the CNN model. The measured GPR (mGPR) values were used as the training data. The CNN model comprises eighteen layers and predicted GPR (pGPR) values. The mGPR and pGPR values were compared, and a cumulative frequency histogram of the prediction error was created to clarify the prediction error tendency. The correlation coefficients of pGPR and mGPR were 0.67, 0.69, 0.66, and 0.73 for 3%/3-mm, 3%/2-mm, 2%/3-mm, and 2%/2-mm gamma criteria, respectively. The respective mean±standard deviations of pGPR-mGPR were -0.87±2.18%, -0.65±2.93%, -0.44±2.53%, and -0.71±3.33%. The probabilities of false positive error cases (pGPR < mGPR) were 72%, 60%, 68%, and 56% for each gamma criterion. We developed a deep learning-based prediction model of the 3D detector array-based VMAT QA for prostate cancer, and evaluated the accuracy and tendency of prediction GPR. This model can provide a proactive estimation for the results of the patient-specific QA before the verification measurement.