Retrospective analysis of CT-guided percutaneous cryoablation for treatment of painful osteolytic bone metastasis

This study aimed to evaluate the safety and effectiveness of CT-guided percutaneous cryoablation for treatment of painful osteolytic bone metastases. A total of 26 patients (36 bone metastases) treated with CT-guided percutaneous cryoablation between May 2012 and June 2016 were enrolled in this retrospective study. All procedures were performed under local anesthesia. A visual analog scale (VAS) was used to evaluate pain before the procedure and at 1 day, 1 month, 3 months, and 6 months after the procedure. Complications during and after the procedure were recorded and graded by the Clavien–Dindo classification. The mean VAS pain score was 7.1 ± 1.1 (range, 4–10) before cryoablation. It was significantly lower at all timepoints after treatment: 2.1 ± 1.7 (P < 0.0001) at 1 day after treatment, 1.3 ± 1.8 (P < 0.0001) at 1 month, 1.6 ± 1.7 (P < 0.0001) at 3 months, and 1.8 ± 1.3 (P < 0.0001) at 6 months. The response rates were 91.7%, 94.4%, 91.7%, and 94.4%, respectively, at 1 day, 1 month, 3 months, and 6 months after cryoablation; the complete response rates were 22.2%, 41.7%, 36.1%, and 22.2%, respectively. Adverse events (skin frostbite, nerve injury, pathologic fracture) occurred in 3 patients. CT-guided percutaneous cryoablation under local anesthesia appears to be a safe and effective treatment for painful osteolytic bone metastases. Prospective clinical trials on large samples needed to confirm this conclusion.     

Impact diagnostique de l’imagerie hybride TEMP/TDM dans la prise en charge des neuroblastomes. Expérience du service de médecine nucléaire du CHU Mohammed VI de Marrakech

IntroductionThe neuroblastoma is a malignant pediatric tumor of the peripheral sympathetic nervous system. This is the pediatric solid extracranial tumor most common and accounts for approximately 8 to 10% of childhood cancers. The diagnosis is based on imaging showing a tumor developed at the expense of the sympathetic nervous system, increased urinary catecholamines, increased uptake of meta-iodo-benzylguanidine (MIBG) and histology who finds a malignant proliferation of small round cells. An assessment of the tumor mass and its extension are required to assess the prognosis and adapt the treatment. The MIBG scintigraphy is a non-invasive imaging technique that can evaluate with a single review the totality of the tumor extension. The single-photon emission computed tomography/computed tomography (SPECT/CT) improves the sensitivity of the examination, it allows an anatomical and functional study and improves the anatomical localization of scintigraphic uptake observed. The purpose of our work is to illustrate the contribution of the SPECT/CT in addition to the planar MIBG scintigraphy in the diagnosis and staging of neuroblastoma about four cases.Case reportClinical case 1: a 4-year old child, presented for 2 months abdominal pain. In the clinical examination, he presented a hard abdominal mass, painless and right paramedian. Abdominal ultrasound showed a right retroperitoneal mass with lymph nodes. The dosage of urinary catecholamines was increased. The ¹³¹I-MIBG scintigraphy showed an image for a right adrenal neuroblastoma measuring 8.6 × 4.5 cm. Surgical excision of the adrenal mass revealed in the anatomopathologic study a malignant tumoral proliferation with round cells compatible with a neuroblastoma. Clinical case 2: a 10-month old infant, presented since 15 days an exophtalmia with poor general status. The clinical examination showed a right abdominal mass and a bilateral periorbital ecchymosis. The radiography of the thorax showed a widening mediastinal. Abdominal echography showed a tissular mass of the right adrenal gland with retroperitoneal lymph nodes and an ascites. The dosage of urinary catecholamines was increased. The ¹³¹I-MIBG scintigraphy showed a right adrenal neuroblastoma measuring 6.4 × 2.7 cm with orbital bone metastases in favor of a Hutchinson syndrome. Clinical case 3: a 2-month old infant, followed since 1 month for bilateral adrenal neuroblastoma. The clinical examination showed an important abdominal distension with bluish nodules under skin. The abdominal echography and the abdomino-pelvic TDM showed two adrenal masses corresponding to a bilateral neuroblastoma with liver metastases. The ¹³¹I-MIBG scintigraphy showed two adrenal masses measuring respectively 6.5 × 3.4 cm and 8 × 6 cm, with liver and skin metastases in favor of 4S neuroblastoma with bilateral adrenal tumors. Clinical case 4: a 3-year old child, followed for left adrenal neuroblastoma with multiple bone metastases. The clinical examination showed a left abdominal mass with exophtalmia and right palpebral ecchymosis. The radiography of the thorax showed a widening of the mediastin with repression of the paravertebral right line. Abdominal echography showed a left retroperitoneal tissular mass measuring 9.2 × 5.2 cm. The abdomino-pelvic TDM showed a left adrenal tumor with lumbar vertebral bone metastases. A first ¹³¹I-MIBG scan showed a left neuroblastoma with multiple bone metastases in the right orbit, the right humerus, the occipital bone, the right scapula and spine (D10, D11, L1, L3, S1). The child was treated by 5 courses of chemotherapy (protocol HRNLB/10). The ¹³¹I-MIBG scan control showed a regression of neuroblastoma size (1.4 × 1.2 cm) with loss of bone metastases of the occipital bone, the right scapula and spine.DiscussionThe MIBG scintigraphy is a simple, non-invasive examination that has excellent sensitivity and specificity in detection of neuroblastoma and especially in invasion bone marrow and in evaluation of the therapeutic response. The hybrid SPECT/CT imaging improve the performance of the scintigraphy as well in sensibility, toward the deep localization, as in specificity for images poorly defined in planar imaging.     

The effect of tumour geometry on the quantification accuracy of planar 123I phantom images

IntroductionAccurate activity quantification is applied in radiation dosimetry. Planar images are important for quantification of whole-body images, enabling assessment of biodistribution from radionuclide administrations. We evaluated the effect of tumour geometry on quantification accuracy of ¹²³I planar phantom studies, including various tumour sizes, tumour-liver distances and two tumour-background ratios.Methods and materialsAn in-house manufactured abdominal phantom was equipped with a liver, different size cylindrical tumours, and a rod for tumour-liver distance variation. The geometric mean method with scatter and attenuation corrections was used for image processing. Scatter and attenuation corrections were made using the triple energy window scatter correction technique and a printed transmission sheet source, respectively. Region definitions for tumour activity distribution compensated for the partial volume effect (PVE). Activity measured in the dose calibrator served as reference for determining quantification accuracy.ResultsThe smallest tumour had the largest percentage deviation with an average activity underestimation of 34.6 ± 1.2%. Activity values for the largest tumour were overestimated by 3.1 ± 3.0%. PVE compensation improved quantification accuracy for all tumour sizes yielding accuracies of <12.4%. Scatter contribution to the tumours from the liver had minimal effect on quantification accuracy at tumour-liver distances >3 cm. With PVE compensation, increased tumour-background ratio resulted in a percentage increase of up to 26.3%.ConclusionWhen applying relevant corrections for scatter, attenuation and PVE without background activity, quantification accuracy of <13% was obtained. We demonstrated the successful implementation of a practical technique to obtain quantitative information from ¹²³I planar images.     

Evaluation of the Block Matching deformable registration algorithm in the field of head-and-neck adaptive radiotherapy

Background and purposeTo compare the accuracy of the Block Matching deformable registration (DIR) against rigid image registration (RIR) for head-and-neck multi-modal images CT to cone-beam CT (CBCT) registration.Material and methodsPlanning-CT and weekly CBCT of 10 patients were used for this study. Several volumes, including medullary canal (MC), thyroid cartilage (TC), hyoid bone (HB) and submandibular gland (SMG) were transposed from CT to CBCT images using either DIR or RIR. Transposed volumes were compared with the manual delineation of these volumes on every CBCT. The parameters of similarity used for analysis were: Dice Similarity Index (DSI), 95%-Hausdorff Distance (95%-HD) and difference of volumes (cc).ResultsWith DIR, the major mean difference of volumes was −1.4 cc for MC, revealing limited under-segmentation. DIR limited variability of DSI and 95%-HD. It significantly improved DSI for TC and HB and 95%-HD for all structures but SMG. With DIR, mean 95%-HD (mm) was 3.01 ± 0.80, 5.33 ± 2.51, 4.99 ± 1.69, 3.07 ± 1.31 for MC, TC, HB and SMG, respectively. With RIR, it was 3.92 ± 1.86, 6.94 ± 3.98, 6.44 ± 3.37 and 3.41 ± 2.25, respectively.ConclusionBlock Matching is a valid algorithm for deformable multi-modal CT to CBCT registration. Values of 95%-HD are useful for ongoing development of its application to the cumulative dose calculation.     

Bone Marrow Uptake in Parathyroid Scintigraphy is A Consequence of Extremely High Parathyroid Hormone Levels

ObjectiveWe retrospectively evaluated patients with end-stage renal disease (ESRD) who were referred to our department for parathyroid scintigraphy. The aim of this study was to investigate the causes of bone marrow uptake observed on parathyroid scintigraphy.MethodsWe included 18 ESRD patients (10 F, 8 M; mean, 52 ± 13 years old; range, 45-59) in the study. The disease duration of the patients was mean 7.7 ± 4.7 years. The patients’ mean plasma calcium and parathormone (PTH) levels were 9.7 ± 1.4 mg/dL and 1,553.3 ± 691.7 pg/mL, respectively. Dual-phase technetium-99m 2-methoxyisobutyl-isonitrile (Tc-99m MIBI) parathyroid imaging and, if necessary, additional Tc-99m pertechnetate scintigraphy were performed. Quantification of the planar early phase parathyroid images was performed for various regions (sternum, humerus, ribs) with the same size rectangular region of interest (ROI, 176 × 176 pixels). Average counts were compared with paired samples Student’s t tests, and P <.05 was considered statistically significant.ResultsTc-99m MIBI parathyroid imaging revealed parathyroid hyperplasia, adenoma, and ectopic adenoma in 7, 3, and 2 patients, respectively. The other 7 patients had normal scintigraphy results with regard to parathyroid pathologies. Bone marrow uptake in the sternum, ribs, and humerus was observed in 6 patients. The difference between the average quantitative value obtained from the ROIs drawn on the sternum and humerus was also statistically significant compared to patients without bone marrow uptake (P<.05). All 6 patients’ exhibited extremely high PTH levels (>2,000 pg/mL; mean, 2,413.7 ± 150 pg/mL) compared to the other 12 patients (mean, 1,342.8 ± 249 pg/mL).ConclusionOur results show that bone marrow uptake on parathyroid scintigraphy is a consequence of extremely high PTH levels in ESRD patients; no further analysis is required. (Endocr Pract. 2013;19:202-205)     

A deep learning classifier for digital breast tomosynthesis

PurposeTo develop a computerized detection system for the automatic classification of the presence/absence of mass lesions in digital breast tomosynthesis (DBT) annotated exams, based on a deep convolutional neural network (DCNN).Materials and MethodsThree DCNN architectures working at image-level (DBT slice) were compared: two state-of-the-art pre-trained DCNN architectures (AlexNet and VGG19) customized through transfer learning, and one developed from scratch (DBT-DCNN). To evaluate these DCNN-based architectures we analysed their classification performance on two different datasets provided by two hospital radiology departments. DBT slice images were processed following normalization, background correction and data augmentation procedures. The accuracy, sensitivity, and area-under-the-curve (AUC) values were evaluated on both datasets, using receiver operating characteristic curves. A Grad-CAM technique was also implemented providing an indication of the lesion position in the DBT slice.Results Accuracy, sensitivity and AUC for the investigated DCNN are in-line with the best performance reported in the field. The DBT-DCNN network developed in this work showed an accuracy and a sensitivity of (90% ± 4%) and (96% ± 3%), respectively, with an AUC as good as 0.89 ± 0.04. A k-fold cross validation test (with k = 4) showed an accuracy of 94.0% ± 0.2%, and a F1-score test provided a value as good as 0.93 ± 0.03. Grad-CAM maps show high activation in correspondence of pixels within the tumour regions.Conclusions We developed a deep learning-based framework (DBT-DCNN) to classify DBT images from clinical exams. We investigated also a possible application of the Grad-CAM technique to identify the lesion position.     

Apport de la scintigraphie au 99mTc-dépréotide pour le diagnostic de lésions osseuses secondaires dans le cancer bronchopulmonaire non à petites cellules stade III–IV

ObjectiveIn non-small cell lung cancer (NSCLC), metastatic bone involvement is usually assessed using conventional ^99mTc-HMDP bone scintigraphy, which has a high sensitivity but a poor specificity. The purpose of this study was to assess the usefulness of the ^99mTc-D scintigraphy for the detection of malignant bone metastases in patients with NSCLC stage III or IV and to compare these results with ^99mTc-HMDP bone scan findings.MethodsNineteen patients (13 M and 6 F, mean age 59 years) with proven NSCLC, suspected to have stage III or IV were enrolled prospectively. All patients underwent whole body ^99mTc-HMDP and ^99mTc-D scintigraphy to detect bone metastases within a mean interval of 14 days. Each focal uptake of ^99mTc-D or ^99mTc-HMDP was considered benign or malignant, leading to positive or negative diagnosis for bone involvement. The final diagnosis of bone metastases was established by a lung specialist, on the basis of additional imaging modalities and of 12 months follow-up.ResultsTwelve bone lesions were identified by ^99mTc-D scintigraphy, 10 were classified as bone metastases and two were classified as inflammatory bone lesions. Four patients were metastatic. Fifty eight bone lesions were detected by ^99mTc-HMDP scintigraphy, 26 of whom were considered malignant, eight patients were thus considered metastatic. Thereby, the two nuclear medicine modalities were concordant for 13 patients, that is 68% of cases and were discordant for six patients, representing 32% of cases. Diagnostic sensitivity, specificity and accuracy of depreotide scintigraphy and ^99mTc-HMDP bone scintigraphy were 75% for both, 93.3 and 73.3%, and 89.5 and 73.3% respectively.ConclusionOur data suggest that depreotide scintigraphy with the same sensitivity, a better accuracy and specificity than those of ^99mTc-HMDP bone scan can detect metastatic bone lesions in patients with NSCLC suspected to have stage III or IV disease.     

基于X线图像的膝关节周围原发性骨肿瘤辅助诊断的机器学习模型研究

摘要 目的：开发机器学习模型,并评估其在膝关节周围原发性骨肿瘤诊断方面的准确性。方法：本文将深度卷积神经网络（DCNN）这一深度学习方法应用于膝关节X线图像的影像组学分析,探讨其辅助诊断膝关节周围原发性骨肿瘤的临床价值。结果：该深度学习模型在区分正常与肿瘤影像方面展现出优异的诊断准确性,使用DCNN模型进行5轮测试的总体准确性为（99.8±0.4）%,而阳性预测值和阴性预测值分别为（100.0±0.0）%和（99.6±0.8）%,各个数据集的曲线下面积（AUC）分别为0.99、1.00、1.00、1.0和1.0,平均AUC为（0.998±0.004）;进一步使用DCNN模型进行了10轮测试显示其在区分良性与恶性骨肿瘤方面的总体准确性为（71.2±1.6）%,且达到了强阳性预测值（91.9±8.5）%,各个数据集的AUC分别为0.63、0.63、0.58、0.69、0.55、0.63、0.54、0.57、0.73、0.63,平均AUC为（0.62±0.06）。结论：本文是首个将人工智能技术应用于骨肿瘤诊断的X线图像影像组学分析方面的研究,人工智能影像组学模型能够帮助医生自动地快速筛查骨肿瘤,确定良性或恶性肿瘤时,阳性预测值较高。     

Regression model-based real-time markerless tumor tracking with fluoroscopic images for hepatocellular carcinoma

PurposeWe have developed a new method to track tumor position using fluoroscopic images, and evaluated it using hepatocellular carcinoma case data.MethodsOur method consists of a training stage and a tracking stage. In the training stage, the model data for the positional relationship between the diaphragm and the tumor are calculated using four-dimensional computed tomography (4DCT) data. The diaphragm is detected along a straight line, which was chosen to avoid 4DCT artifact. In the tracking stage, the tumor position on the fluoroscopic images is calculated by applying the model to the diaphragm. Using data from seven liver cases, we evaluated four metrics: diaphragm edge detection error, modeling error, patient setup error, and tumor tracking error. We measured tumor tracking error for the 15 fluoroscopic sequences from the cases and recorded the computation time.ResultsThe mean positional error in diaphragm tracking was 0.57 ± 0.62 mm. The mean positional error in tumor tracking in three-dimensional (3D) space was 0.63 ± 0.30 mm by modeling error, and 0.81–2.37 mm with 1–2 mm setup error. The mean positional error in tumor tracking in the fluoroscopy sequences was 1.30 ± 0.54 mm and the mean computation time was 69.0 ± 4.6 ms and 23.2 ± 1.3 ms per frame for the training and tracking stages, respectively.ConclusionsOur markerless tracking method successfully estimated tumor positions. We believe our results will be useful in increasing treatment accuracy for liver cases.     

Comparison of 68GA-FAPI-04 PET/CT and 18F-FDG PET/CT in detection of metastatic bone disease in various cancers

ObjectiveOur aim in this retrospective study was to compare the diagnostic accuracy of ⁶⁸Ga-FAPI-04 PET/CT and ¹⁸F-FDG PET/CT in detecting bone metastases of various cancers and to evaluate the potential usefulness of ⁶⁸Ga-FAPI-04 PET/CT in detecting metastatic bone disease.Material and methodOur retrospective study included 44 patients diagnosed with bone metastases due to various cancers between January 2021 and February 2022. All patients underwent ⁶⁸Ga-FAPI-04 PET/CT and ¹⁸F-FDG PET/CT imaging within 14 days. In the semi-quantitative analysis of the skeletal system, all regions with higher uptake than background activity were considered pathological. SUVmax and Metastasis-to-background ratio (TBR) values were calculated from metastatic sites.ResultsA total of 827 bone metastases were detected in our study. The diagnostic accuracies of FAPI PET/CT and ¹⁸F-FDG PET/CT were 91.8% and 81.5%, respectively (P < 0.001). When all bone metastases were compared, the SUVmax of ⁶⁸Ga-FAPI-04 PET/CT was statistically significantly higher than that of ¹⁸F-FDG PET/CT (median 6.15 vs. 5.2; P < 0.001). When FDG and FAPI SUVmax values were compared according to metastasis types, FAPI SUVmax and TBR values in osteolytic, medullary and mixed type bone metastases were found to be statistically significantly higher than FDG (P-values: < 0.001, < 0.001, < 0.001, respectively). There was no statistically significant difference between FDG and FAPI SUVmax values in osteoblastic bone metastases (P = 0.26).ConclusionIt has been shown that ⁶⁸Ga-FAPI-04 PET/CT is superior to ¹⁸F-FDG PET/CT in detecting metastatic bone disease and may have more clinical impact on disease management.     

Simple low-cost approaches to semantic segmentation in radiation therapy planning for prostate cancer using deep learning with non-contrast planning CT images

PurposeDeep learning has shown great efficacy for semantic segmentation. However, there are difficulties in the collection, labeling and management of medical imaging data, because of ethical complications and the limited number of imaging studies available at a single facility.This study aimed to find a simple and low-cost method to increase the accuracy of deep learning semantic segmentation for radiation therapy of prostate cancer.MethodsIn total, 556 cases with non-contrast CT images for prostate cancer radiation therapy were examined using a two-dimensional U-Net. Initially, all slices were used for the input data. Then, we removed slices of the cranial portions, which were beyond the margins of the bladder and rectum. Finally, the ground truth labels for the bladder and rectum were added as channels to the input for the prostate training dataset.ResultsThe highest mean dice similarity coefficients (DSCs) for each organ in the test dataset of 56 cases were 0.85 ± 0.05, 0.94 ± 0.04 and 0.85 ± 0.07 for the prostate, bladder and rectum, respectively. Removal of the cranial slices from the original images significantly increased the DSC of the rectum from 0.83 ± 0.09 to 0.85 ± 0.07 (p < 0.05). Adding bladder and rectum information to prostate training without removing the slices significantly increased the DSC of the prostate from 0.79 ± 0.05 to 0.85 ± 0.05 (p < 0.05).ConclusionsThese cost-free approaches may be useful for new applications, which may include updated models and datasets. They may be applicable to other organs at risk (OARs) and clinical targets such as elective nodal irradiation.     

Respiratory motion of adrenal gland metastases: Analyses using four-dimensional computed tomography images

PurposeTo evaluate the respiratory motion of adrenal gland metastases in three-dimensional directions using four-dimensional computed tomography (4DCT) images.MethodsFrom January 2013 to May 2016, 12 patients with adrenal gland metastases were included in this study. They all underwent 4DCT scans to assess respiratory motion of adrenal gland metastases in free breathing state. The 4DCT images were sorted into 10 image series according to the respiratory phase from the end inspiration to the end expiration, and then transferred to FocalSim workstation. All gross tumor volumes (GTVs) of adrenal gland metastases were drawn by a single physician and confirmed by a second. Relative coordinates of adrenal gland metastases were automatically generated to calculate adrenal gland metastases motion in different axial directions.ResultsThe average respiratory motion of adrenal gland metastases in left-right (LR), cranial-caudal (CC), anterior-posterior (AP), 3-dimensional (3D) vector directions was 3.4 ± 2.2 mm, 9.5 ± 5.5 mm, 3.8 ± 2.0 mm and 11.3 ± 5.3 mm, respectively. The ratios were 58.6% ± 11.4% and 63.2% ± 12.5% when the volumes of GTV_In0% and GTV _In100% were compared with volume of IGTV_10phase. The volume ratio of IGTV_10phase to GTV_3D was 1.73 ± 0.48.ConclusionsAdrenal gland metastasis is a respiration-induced moving target, and an internal target volume boundary should be provided when designing the treatment plan. The CC motion of adrenal gland metastasis is predominant and >5 mm, thus motion management strategies are recommended for patients undergoing external radiotherapy for adrenal gland metastasis.     

Comparison of different deep learning architectures for synthetic CT generation from MR images

PurposeAmong the different available methods for synthetic CT generation from MR images for the task of MR-guided radiation planning, the deep learning algorithms have and do outperform their conventional counterparts. In this study, we investigated the performance of some most popular deep learning architectures including eCNN, U-Net, GAN, V-Net, and Res-Net for the task of sCT generation. As a baseline, an atlas-based method is implemented to which the results of the deep learning-based model are compared.MethodsA dataset consisting of 20 co-registered MR-CT pairs of the male pelvis is applied to assess the different sCT production methods' performance. The mean error (ME), mean absolute error (MAE), Pearson correlation coefficient (PCC), structural similarity index (SSIM), and peak signal-to-noise ratio (PSNR) metrics were computed between the estimated sCT and the ground truth (reference) CT images.ResultsThe visual inspection revealed that the sCTs produced by eCNN, V-Net, and ResNet, unlike the other methods, were less noisy and greatly resemble the ground truth CT image. In the whole pelvis region, the eCNN yielded the lowest MAE (26.03 ± 8.85 HU) and ME (0.82 ± 7.06 HU), and the highest PCC metrics were yielded by the eCNN (0.93 ± 0.05) and ResNet (0.91 ± 0.02) methods. The ResNet model had the highest PSNR of 29.38 ± 1.75 among all models. In terms of the Dice similarity coefficient, the eCNN method revealed superior performance in major tissue identification (air, bone, and soft tissue).ConclusionsAll in all, the eCNN and ResNet deep learning methods revealed acceptable performance with clinically tolerable quantification errors.     

Recognition of brain tumors in MRI images using texture analysis

ObjectivesBrain neoplasms or intracranial tumors, which are more common in older adults, can affect individuals of any age including pediatric and children. Exposure to carcinogenic agents including ionizing radiation and family history is one of the main causes of the disease. Early diagnosis is crucial to avoid prolonged. patients' suffering. The aim of the study was to efficiently recognize the brain tumors from the other brain tissues which include grey and white matter as well as cerebrospinal fluid (CSF).Materials and methodsThis study was performed using axial, sagittal and coronal views for fifty brain tumor patients randomly selected from a set of 200 patients, with a “control” set consisting of images showing no sign of disease; and the “test” brain MRI images for patients diagnosed with brain tumor. The study includes both genders with age ranging from 18 years to 83 years old, (56.5 ± 17.2). The brain images were acquired using a standard head coil Philips Intera 1.5 Tesla machine (USA). The thickness of each section in the entire sequence was 8 mm. Acquisition of T2-weighted and T1-weighted were performed. Interactive Data Language software was used to analyze the data.ResultsThe results of this study showed that: the overall accuracy of classification process was 94.8%, and for the tumor; the sensitivity was 97.3%. White matter and grey matter showed a classification accuracy of 95.7% and 89.7% and for CSF the accuracy was 94.3%.ConclusionThe results showed that brain tumor can be classified successfully and delineated using texture analysis with minimum efforts and with high accuracy for brain tumors.     

基于对象特征的山东省丘陵地区多时相遥感土地覆被自动分类

对于基于像元的土地覆被分类来说,植被的分类是难点。使用多时相面向对象分类方法可以较好的解决这个问题。以山东省烟台市丘陵地区为研究区,采用Landsat TM(Landsat Thematic Mapper remotely sensed imagery)、DEM(Digital Elevation Model)、坡度、坡位、坡向等多种数据,利用基于对象特征的多时相分类方法对研究区进行土地覆盖自动分类。首先对影像进行多尺度分割并检验分割结果选取合适的分割尺度,然后分析对象的光谱、纹理、形状特征。根据各类地物的光谱特征、地理相关性、形状、空间分布等特征,明确类别之间的差异。建立决策树使用隶属度函数进行模糊分类,借助支持向量机提高分类精度。研究结果表明,通过使用多时相影像采用面向对象分类方法,相对于传统的基于像素的分类可以明显提高分类精度,尤其是解决了乔灌草的区分问题。     

A Deep Learning-based correction to EPID dosimetry for attenuation and scatter in the Unity MR-Linac system

PurposeEPID dosimetry in the Unity MR-Linac system allows for reconstruction of absolute dose distributions within the patient geometry. Dose reconstruction is accurate for the parts of the beam arriving at the EPID through the MRI central unattenuated region, free of gradient coils, resulting in a maximum field size of ~10 × 22 cm² at isocentre. The purpose of this study is to develop a Deep Learning-based method to improve the accuracy of 2D EPID reconstructed dose distributions outside this central region, accounting for the effects of the extra attenuation and scatter.MethodsA U-Net was trained to correct EPID dose images calculated at the isocenter inside a cylindrical phantom using the corresponding TPS dose images as ground truth for training. The model was evaluated using a 5-fold cross validation procedure. The clinical validity of the U-Net corrected dose images (the so-called DEEPID dose images) was assessed with in vivo verification data of 45 large rectum IMRT fields. The sensitivity of DEEPID to leaf bank position errors (±1.5 mm) and ±5% MU delivery errors was also tested.ResultsCompared to the TPS, in vivo 2D DEEPID dose images showed an average γ-pass rate of 90.2% (72.6%–99.4%) outside the central unattenuated region. Without DEEPID correction, this number was 44.5% (4.0%–78.4%). DEEPID correctly detected the introduced delivery errors.ConclusionsDEEPID allows for accurate dose reconstruction using the entire EPID image, thus enabling dosimetric verification for field sizes up to ~19 × 22 cm² at isocentre. The method can be used to detect clinically relevant errors.     

Enhancement of Hounsfield unit distribution in cone-beam CT images for adaptive radiation therapy: Evaluation of a hybrid correction approach

PurposeThis study aims to evaluate the accuracy of a hybrid approach combining the histogram matching (HM) and the multilevel threshold (MLT) to correct the Hounsfield Unit (HU) distribution in cone-beam CT (CBCT) images.Methods and MaterialsCBCT images acquired for ten prostate cancer patients were processed by matching their histograms to those of deformed planning CT (pCT) images obtained after applying a deformable registration (DR) process. Then, HU values corresponding to five tissue types in the pCT were assigned to the obtained CBCT images (CBCT_HM-MLT). Finally, the CBCT_HM-MLT images were compared to the deformed pCT visually and using different statistical metrics.ResultsThe visual assessment and the profiles comparison showed that the high discrepancies in the CBCT images were significantly reduced when using the proposed approach. Furthermore, the correlation values indicated that the CBCT_HM-MLT were in good agreement with the deformed pCT with correlation values ranging from 0.9893 to 0.9962. In addition, the root mean squared error (RMSE) over the entire volume was reduced from 64.15 ± 9.50 to 51.20 ± 6.76 HU. Similarly, the mean absolute error in specific tissue classes was significantly reduced especially in the soft tissue-air interfaces. These results confirmed that applying MLT after HM worked better than using only HM for which the correlation values were ranging from 0.9878 to 0.9955 and the RMSE was 55.95 ± 10.43 HU.ConclusionEvaluation of the proposed approach showed that the HM + MLT correction can improve the HU distribution in the CBCT images and generate corrected images in good agreement with the pCT.     

Dual-energy contrast-enhanced digital mammography: Glandular dose estimation using a Monte Carlo code and voxel phantom

PurposeTo estimate the mean glandular dose of contrast enhanced digital mammography, using the EGSnrc Monte Carlo code and female adult voxel phantom.MethodsAutomatic exposure control of full field digital mammography system was used for the selection of the X-ray spectrum and the exposure settings for dual energy imaging. Measurements of the air-kerma and of the half value layers were performed and a Monte Carlo simulation of the digital mammography system was used to compute the mean glandular dose, for breast phantoms of various thicknesses, glandularities and for different X-ray spectra (low and high energy).ResultsFor breast phantoms of 2.0–8.0 cm thick and 0.1–100% glandular fraction, CC view acquisition, from AEC settings, can result in a mean glandular dose of 0.450 ± 0.022 mGy −2.575 ± 0.033 mGy for low energy images and 0.061 ± 0.021 mGy – 0.232 ± 0.033 mGy for high energy images. In MLO view acquisition mean glandular dose values ranged between 0.488 ± 0.007 mGy – 2.080 ± 0.021 mGy for low energy images and 0.065 ± 0.012 mGy – 0.215 ± 0.010 mGy for high energy images.ConclusionThe low kV part of contrast enhanced digital mammography is the main contributor to total mean glandular breast dose. The results of this study can be used to provide an estimated mean glandular dose for individual cases.     

Glioblastomas and brain metastases differentiation following an MRI texture analysis-based radiomics approach

PurposeTo evaluate the potential of 2D texture features extracted from magnetic resonance (MR) images for differentiating brain metastasis (BM) and glioblastomas (GBM) following a radiomics approach.MethodsThis retrospective study included 50 patients with BM and 50 with GBM who underwent T1-weighted MRI between December 2010 and January 2017. Eighty-eight rotation-invariant texture features were computed for each segmented lesion using six texture analysis methods. These features were also extracted from the four images obtained after applying the discrete wavelet transform (88 features × 4 images). Three feature selection methods and five predictive models were evaluated. A 5-fold cross-validation scheme was used to randomly split the study group into training (80 patients) and testing (20 patients), repeating the process ten times. Classification was evaluated computing the average area under the receiver operating characteristic curve. Sensibility, specificity and accuracy were also computed. The whole process was tested quantizing the images with different gray-level values to evaluate their influence in the final results.ResultsHighest classification accuracy was obtained using the original images quantized with 128 gray-levels and a feature selection method based on the p-value. The best overall performance was achieved using a support vector machine model with a subset of 32 features (AUC = 0.896 ± 0.067, sensitivity of 82% and specificity of 80%). Naïve Bayes and k-nearest neighbors models showed also valuable results (AUC ≈ 0.8) with a lower number of features (<13), thus suggesting that these models may be more generalizable when using external validations.ConclusionThe proposed radiomics MRI approach is able to discriminate between GBM and BM with high accuracy employing a set of 2D texture features, thus helping in the diagnosis of brain lesions in a fast and non-invasive way.     

18F-sodium fluoride PET/CT indications

¹⁸F-sodium fluoride uptake mechanism is based on the fluoride ion exchange, with a hydroxyl group of the hydroxyapatite being at the surface of the bone crystals and thus forming fluorapatite, especially at sites with a high rate of bone remodeling. ¹⁸F-fluoride shows a lower protein binding and faster plasma clearance with respect to ^99mTc-diphosphonates (twice increased bone uptake). The adult dose range is 185–370 MBq, whereas in children it is advisable to adjust dose (0.057 mCi/kg). The acquisition time of the ¹⁸F-fluoride PET study is similar to that of ¹⁸F-FDG one (60 ± 30 minutes). The CT is required for attenuation correction (low amperage 30 mAs) but also allows artifact reduction and quantification. ¹⁸F-fluoride PET shows high sensitivity for detection of bone metastases, affecting its specificity. However, CT addition may increase specificity, resulting in a high accuracy of ¹⁸F-fluoride PET/CT. Given its uptake mechanism, its sensitivity is higher for mixed and blastic metastases rather than for lytic ones. Multiple studies have shown that the sensitivity of the ¹⁸F-fluoride is higher than bone scan, including whole-body bone SPECT. However, ¹⁸F-fluoride PET/CT should find their real indications in the current algorithms of clinical practice.