Performance comparison of different medical image fusion algorithms for clinical glioma grade classification with advanced magnetic resonance imaging (MRI).

Non-invasive glioma grade classification is an exciting area in neuroimaging. The primary purpose of this study is to investigate the performance of different medical image fusion algorithms for glioma grading purposes by fusing advanced Magnetic Resonance Imaging (MRI) images. Ninety-six subjects underwent an Apparent diffusion coefficient (ADC) map and Susceptibility-weighted imaging (SWI) MRI scan. After preprocessing, the different medical image fusion methods used to fuse ADC maps and SWI were Principal Component Analysis (PCA), Structure-Aware, Discrete Cosine Harmonic Wavelet Transform (DCHWT), Deep-Convolutional Neural network (DNN), Dual-Discriminator conditional generative adversarial network (DDcGAN), and Laplacian Re-Decomposition (LRD). The Entropy, standard deviation (STD), peak signal-to-noise ratio (PSNR), structural similarity index measure (SSIM), and Relative Signal Contrast (RSC) were calculated for qualitative and quantitative analysis. We found high fused image quality with LRD and DDcGAN methods. Further quantitative analysis showed that RSCs in fused images in Low-Grade glioma (LGG) were significantly higher than RSCs in High-Grade glioma (HGG) with PCA, DCHWT, LRD, and DDcGAN. The Receiver Operating Characteristic (ROC) curve test highlighted that LRD and DDcGAN have the highest performance for glioma grade classification. Our work suggests using the DDcGAN and LRD networks for glioma grade classification by fusing ADC maps and SWI images.

Radiomics based predictive modeling of rectal toxicity in prostate cancer patients undergoing radiotherapy: CT and MRI comparison.

BACKGROUND: Rectal toxicity is one of the common side effects after radiotherapy in prostate cancer patients. Radiomics is a non-invasive and low-cost method for developing models of predicting radiation toxicity that does not have the limitations of previous methods. These models have been developed using individual patients' information and have reliable and acceptable performance. This study was conducted by evaluating the radiomic features of computed tomography (CT) and magnetic resonance (MR) images and using machine learning (ML) methods to predict radiation-induced rectal toxicity. METHODS: Seventy men with pathologically confirmed prostate cancer, eligible for three-dimensional radiation therapy (3DCRT) participated in this prospective trial. Rectal wall CT and MR images were used to extract first-order, shape-based, and textural features. The least absolute shrinkage and selection operator (LASSO) was used for feature selection. Classifiers such as Random Forest (RF), Decision Tree (DT), Logistic Regression (LR), and K-Nearest Neighbors (KNN) were used to create models based on radiomic, dosimetric, and clinical data alone or in combination. The area under the curve (AUC) of the receiver operating characteristic curve (ROC), accuracy, sensitivity, and specificity were used to assess each model's performance. RESULTS: The best outcomes were achieved by the radiomic features of MR images in conjunction with clinical and dosimetric data, with a mean of AUC: 0.79, accuracy: 77.75%, specificity: 82.15%, and sensitivity: 67%. CONCLUSIONS: This research showed that as radiomic signatures for predicting radiation-induced rectal toxicity, MR images outperform CT images.

Ultrasound-guided chemoradiotherapy of breast cancer using smart methotrexate-loaded perfluorohexane nanodroplets.

Chemoradiotherapy with controlled-release nanocarriers such as sono-sensitive nanodroplets (NDs) can enhance the anticancer activity of chemotherapy medicines and reduces normal tissue side effects. In this study, folic acid-functionalized methotrexate-loaded perfluorohexane NDs with alginate shell (FA-MTX/PFH@alginate NDs) were synthesized, characterized, and their potential for ultrasound-guided chemoradiotherapy of breast cancer was investigated in vitro and in vivo. The cancer cell (4T1) viabilities and surviving fractions after NDs and ultrasound treatments were significantly decreased. However, this reduction was much more significant for ultrasound in combination with X-ray irradiation. The in vitro and in vivo results confirmed that MTX-loaded NDs are highly biocompatible and they have no significant hemolytic activity and organ toxicity. Furthermore, the in vivo results indicated that the FA-MTX/PFH@alginate NDs were accumulated selectively in the tumor region. In conclusion, FA-functionalized MTX/PFH@alginate NDs have a great theranostic performance for ultrasound-controlled drug delivery in combination with radiotherapy of breast cancer.

Multiparametric study for glioma grading with FLAIR, ADC map, eADC map, T1 map, and SWI images.

PURPOSE: This paper is a preliminary attempt to compare the diagnostic efficiency and performance of Fluid-attenuated inversion recovery (FLAIR), apparent diffusion coefficient (ADC) map, exponential ADC (eADC) map, T1 map, and Susceptibility-weighted image (SWI) for glioma grading and combine these image data pairs to compare the diagnostic performance of different image data pairs for glioma grading. MATERIAL AND METHODS: Fifty-nine patients underwent FLAIR, ADC map, eADC map, Variable flip-angle (VFA) spoiled gradient recalled echo (SPGR) method, and SWI MRI imaging. The T1 map was reconstructed by the VFA-SPGR method. The average Relative Signal Contrast (RSC) and receiver operating characteristic curve (ROC) was calculated in a different image. The multivariate binary logistic regression model combined different image data pairs. RESULTS: The average RSC of SWI and ADC maps in high-grade glioma is significantly lower than RSCs in low-grade. The average RSC of the eADC map and T1 maps increased with glioma grade. No significant difference was detected between low and high-grade glioma on FLAIR images. The AUC for low and high-grade glioma differentiation on ADC maps, eADC maps, T1 map, and SWI were calculated 0.781, 0.864, 0.942, and 0.904, respectively. Also, by adding different image data, diagnostic performance was increased. CONCLUSION: Interestingly, the T1 map and SWI image have the potential to use in the clinic for glioma grading purposes due to their high performance. Also, the eADC map+T1 map and T1 map+SWI image weights have the highest diagnostic performance for glioma grading.

Glioma segmentation with DWI weighted images, conventional anatomical images, and post-contrast enhancement magnetic resonance imaging images by U-Net.

Glioma segmentation is believed to be one of the most important stages of treatment management. Recent developments in magnetic resonance imaging (MRI) protocols have led to a renewed interest in using automatic glioma segmentation with different MRI image weights. U-Net is a major area of interest within the field of automatic glioma segmentation. This paper examines the impact of different input MRI image-weight on the U-Net output performance for glioma segmentation. One hundred forty-nine glioma patients were scanned with a 1.5T MRI scanner. The main MRI image-weights acquired are diffusion-weighted imaging (DWI) weighted images (b50, b500, b1000, Apparent diffusion coefficient (ADC) map, Exponential apparent diffusion coefficient (eADC) map), anatomical image-weights (T2, T1, T2-FLAIR), and post enhancement image-weights (T1Gd). The U-Net and data augmentation are used to segment the glioma tumors. Having the Dice coefficient and accuracy enabled us to compare our results with the previous study. The first set of analyses examined the impact of epoch number on the accuracy of U-Net, and n_epoch = 20 was selected for U-Net training. The mean Dice coefficient for b50, b500, b1000, ADC map, eADC map, T2, T1, T2-FLAIR, and T1Gd image weights for glioma segmentation with U-Net were calculated 0.892, 0.872, 0.752, 0.931, 0.944, 0.762, 0.721, 0.896, 0.694 respectively. This study has found that, DWI image-weights have a higher diagnostic value for glioma segmentation with U-Net in comparison with anatomical image-weights and post enhancement image-weights. The results of this investigation show that ADC and eADC maps have higher performance for glioma segmentation with U-Net.

Preliminary study of multiple b-value diffusion-weighted images and T1 post enhancement magnetic resonance imaging images fusion with Laplacian Re-decomposition (LRD) medical fusion algorithm for glioma grading.

BACKGROUND: Grade of brain tumor is thought to be the most significant and crucial component in treatment management. Recent development in medical imaging techniques have led to the introduce non-invasive methods for brain tumor grading such as different magnetic resonance imaging (MRI) protocols. Combination of different MRI protocols with fusion algorithms for tumor grading is used to increase diagnostic improvement. This paper investigated the efficiency of the Laplacian Re-decomposition (LRD) fusion algorithms for glioma grading. PROCEDURES: In this study, 69 patients were examined with MRI. The T1 post enhancement (T1Gd) and diffusion-weighted images (DWI) were obtained. To evaluated LRD performance for glioma grading, we compared the parameters of the receiver operating characteristic (ROC) curves. FINDINGS: We found that the average Relative Signal Contrast (RSC) for high-grade gliomas is greater than RSCs for low-grade gliomas in T1Gd images and all fused images. No significant difference in RSCs of DWI images was observed between low-grade and high-grade gliomas. However, a significant RSCs difference was detected between grade III and IV in the T1Gd, b50, and all fussed images. CONCLUSIONS: This research suggests that T1Gd images are an appropriate imaging protocol for separating low-grade and high-grade gliomas. According to the findings of this study, we may use the LRD fusion algorithm to increase the diagnostic value of T1Gd and DWI picture for grades III and IV glioma distinction. In conclusion, this article has emphasized the significance of the LRD fusion algorithm as a tool for differentiating grade III and IV gliomas.

Folic acid-functionalized gadolinium-loaded phase transition nanodroplets for dual-modal ultrasound/magnetic resonance imaging of hepatocellular carcinoma.

Dual-modal molecular imaging by combining two imaging techniques can provide complementary information for early cancer diagnosis and therapeutic monitoring. In the present manuscript, folic acid (FA)-functionalized gadolinium-loaded nanodroplets (NDs) are introduced as dual-modal ultrasound (US)/magnetic resonance (MR) imaging contrast agents. These phase-change contrast agents (PCCAs) with alginate (Alg) stabilizing shell and a liquid perfluorohexane (PFH) core were successfully synthesized via the nano-emulsion method and characterized. In this regard, mouse hepatocellular carcinoma (Hepa1-6) as target cancer cells and mouse fibroblast (L929) as control cells were used. The in vitro and in vivo cytotoxicity assessments indicated that Gd/PFH@Alg and Gd/PFH@Alg-FA nanodroplets are highly biocompatible. Gd-loaded NDs do not induce organ toxicity, and no significant hemolytic activity in human red blood cells is observed. Additionally, nanodroplets exhibited strong ultrasound signal intensities as well as T1-weighted MRI signal enhancement with a high relaxivity value of 6.40 mM-1 s-1, which is significantly higher than that of the clinical Gadovist contrast agent (r1 = 4.01 mM-1 s-1). Cellular uptake of Gd-NDs-FA by Hepa1-6 cancer cells was approximately 2.5-fold higher than that of Gd-NDs after 12 h incubation. Furthermore, in vivo results confirmed that the Gd-NDs-FA bound selectively to cancer cells and were accumulated in the tumor region. In conclusion, Gd/PFH@Alg-FA nanodroplets have great potential as US/MR dual-modal imaging nanoprobes for the early diagnosis of cancer.

Trastuzumab and folic acid functionalized gold nanoclusters as a dual-targeted radiosensitizer for megavoltage radiation therapy of human breast cancer.

In the present study, the effect of functionalized gold nanoclusters (AuNCs) with trastuzumab (Herceptin®) and/or folic acid (FA) as a single and dual-targeted radiosensitizers for the enhancement of megavoltage radiation therapy efficacy was investigated. SK-BR3 breast cancer cells as human epidermal growth factor 2 (HER2) and folate overexpressing cell line and the murine fibroblast (L929) as a control cell line were selected. The cellular uptake was followed using inductively coupled plasma optical emission spectrometry (ICP-OES) that showed AuNCs-FA-HER uptake by SK-BR3 cells was 3 times more than the non-targeted AuNCs after 12 h incubation. MTT and clonogenic assays revealed that the viability and surviving fraction of cancer cells were significantly inhibited by treating with all AuNCs under radiation compared to treating with radiation alone. However, these effects in the dual-targeted AuNCs group (AuNCs-FA-HER) was significantly greater than non-targeted and single-targeted AuNCs groups. Also, apoptosis was evaluated using an Annexin V-FITC/propidium iodide (PI) kit in flow cytometry. All AuNCs, in combination with 4 Gy of photon beam, induced more apoptosis. By fitting the survival fraction data on the linear-quadratic model, the sensitization enhancement factor (SER) of AuNCs, AuNCs-FA, AuNCs-HER, and AuNCs-FA-HER, were obtained 1.17, 1.32, 1.48 and 1.77, respectively. SER for AuNCs-FA-HER was significantly higher than that non-targeted and single-targeted AuNCs (p-value < 0.05) that can be attributed to more internalization in the cancer cells. It was concluded that functionalized AuNCs with both folic acid and Herceptin could represent a promising strategy for increased cellular internalization that improved radiation therapy efficiency in SK-BR3 breast cancer cells.

Mn-doped ZnS quantum dots-chlorin e6 shows potential as a treatment for chondrosarcoma: an in vitro study.

Chondrosarcoma is the second-most malignant cancer of the bone and routine treatments such as chemotherapy and radiotherapy have not responded to the treatment of this cancer. Due to the resistance of chondrosarcoma to radiotherapy, the combination of therapeutic methods has been considered in recent years. In this study, a novel combination approach is used that allows photodynamic therapy to be activated by X-rays. The synthesis of Mn-doped zinc sulphide (ZnS) quantum dots was carried out and chlorin e6 photosensitiser attached by covalent and non-covalent methods and their application as an intracellular light source for photodynamic activation was investigated. The toxicity of each nanoparticles was evaluated on chondrosarcoma cancer cells (SW1353) before and after radiation. Also, the effect nanoparticle-photosensitiser conjugated type was investigated in the therapeutic efficacy. The characterisation test (SEM, TEM, EDS, TGA, XRD and ICP analyses) was shown successful synthesis of Mn-doped ZnS quantum dots. Chondrosarcoma cancer cell viability was significantly reduced when cells were treated with MPA-capped Mn-doped ZnS quantum dots-chlorin e6 with spermine linker and with covalent attachment (P ≤ 0.001). These results indicate that X-ray can activate the quantum dot complexes for cancer treatment, which can be a novel method for treatment of chondrosarcoma.

Could FA-PG-SPIONs act as a hyperthermia sensitizing agent? An in vitro study.

The therapeutic effect of polyglycerol coated iron oxide nanoparticles (PG-SPIONs, non-targeted nanoparticles) and folic acid-conjugated polyglycerol coated iron oxide nanoparticles (FA-PG-SPIONs, targeted nanoparticles) in combination with hyperthermia on viability of HeLa cells was investigated. It was observed that coated and uncoated SPIONs have spherical shapes with an average diameter of 17.9 ±â€¯2.85 nm and 5.4 ±â€¯0.75 nm, respectively. The penetration rate for cells treated with targeted nanoparticles was shown to be more than that of non-targeted nanoparticles. Moreover, it was revealed that the treatment of cells with ≥ 50 µg/ml FA-PG-SPIONs in combination with hyperthermia induced cytotoxicity in comparison to control cells. The results also showed that increasing the concentrations of targeted nanoparticles (FA-PG-SPIONs) and heating time would increase the value of thermal enhancement factor (TEF). In contrast, TEF values were not increased with increasing heating time and concentrations of non-targeted nanoparticles (PG-SPIONs). On the other hand, TEF values were increased with increasing concentrations and heating time so that the maximum TEF value was obtained at the highest concentration (FA-PG-SPION, 200 µg/ml) as well as the longest heating duration (60 min). Thus, it is concluded that FA-PG-SPIONs with concentrations ≥ 100 µg/ml could be introduced and used as hyperthermia sensitizing agents leading to enhanced cancer therapy efficiency.

Evaluating the effect of the vacuum bag on the dose distribution in radiation therapy.

INTRODUCTION: Immobilization of patients in radiation therapy can be performed with a vacuum bag (VB). The aim of this study is to measure the effect of the VB in the surface and depth dose of patients in radiation therapy. MATERIALS AND METHODS: The effect of the VB on the surface dose and depth dose is measured in clinical conditions. Various dosimeters were used in following measurements: parallel plate chamber for depth dose, farmer ionization chamber for various gantry angles, and Mapcheck2 dosimeter for various thicknesses of VB. The effect of VB cap, which may be placed in the beam path, is also measured using EDR2 film. The measurements were performed for 6 MV and 18 MV photons with an Oncor linac. RESULTS: The increase of 30% and 25% in the surface dose with VB was observed for 6 MV and 18 MV, respectively. Though due to the use of VB, the reduction of the absorbed dose at a 5 cm depth is under 1% and can be ignored in MU calculation. For various thicknesses of VB, 8-14 cm, the attenuation of the primary beam were up to 2.5% for 6 MV and 1.2% for 18 MV photon. The presence of VB cap in the path of radiation reduced the depth dose up to 15% and 11% for 6 MV and 18 MV, respectively. CONCLUSION: The use of VB can increase the surface dose of the patient up to 30% and this fact should be considered in treatment planning. For some lateral fields the cap of the VB may interfere with radiation field. If the cap of VB is placed in the beam path, it can cause a cold spot in tumor. DISCUSSION: The use of VB can increase the surface dose of the patient up to 30%. For some lateral fields, the cap of the VB might interfere with the radiation field. If the cap of VB is placed in the beam path, it can cause a cold spot in the tumor volume.

The role of folic acid-conjugated polyglycerol coated iron oxide nanoparticles on radiosensitivity with clinical electron beam (6 MeV) on human cervical carcinoma cell line: In vitro study.

BACKGROUND AND PURPOSE: The objective of this study was to investigate the therapeutic effect of Folic Acid-Conjugated polyglycerol coated iron oxide nanoparticles on the radiosensitivity of HeLa cells when irradiated with 6 MeV electron beams. MATERIALS AND METHODS: Different concentrations of iron oxide nanoparticles (PG-SPIONs and FA-PG-SPIONs (25, 50, 100, 200 µg ml-1)) were synthesized by the thermal decomposition technique. The effect of PG-SPIONs and FA-PG-SPIONs in combination with radiation (2, 4, 6 Gy) on the viability of cells and cell survival were estimated using the trypan blue dye exclusion test and MTT assay immediately and 48 h after irradiations, respectively. RESULTS: It was observed that the penetration rate of uptake for cells treated with >50 µg ml-1 FA-PG-SPIONs was more than that of non-targeted nanoparticles. The data obtained by trypan blue dye exclusion test showed no significant reduction in cell viability for all groups in comparison with control group. The results revealed that increasing the radiation doses in the presence of the concentrations of the nanoparticles increased the value of radiosensitivity. The most radiosensitivity was obtained at the highest concentration of FA-PG-SPIONs (200 µg ml-1) as well as the longest radiation doses. CONCLUSION: It was revealed that higher concentrations of the FA-PG-SPIONs in combination with 6 MeV electron beams could enhance radiosensitization of HeLa cells.

Dosimetry and evaluating the effect of treatment parameters on the leakage of multi leaf collimators in ONCOR linear accelerators.

BACKGROUND: One of the standard equipment in medical linear accelerators is multi-leaf collimators (MLCs); which is used as a replacement for lead shielding. MLC's advantages are a reduction of the treatment time, the simplicity of treatment, and better dose distribution. The main disadvantage of MLC is the radiation leakages from the edges and between the leaves. The purpose of this study was to determine the effect of various treatment parameters in the magnitude of MLC leakage in linear accelerators. MATERIALS AND METHODS: This project was performed with ONCOR Siemens linear accelerators. The amount of radiation leakage was determined by film dosimetry method. The films were Kodak-extended dose range-2, and the beams were 6 MV and 18 MV photons. In another part of the experiment, the fluctuation of the leakage was measured at various depths and fields. RESULTS: The amount of leakage was generally up to 1.5 ± 0.2% for both energies. The results showed that the level of the leakage and the amount of dose fluctuation depends on the field size and depth of measurement. The amount of the leakage fluctuations in all energies was decreased with increasing of field size. The variation of the leakage versus field size was similar to the inverse of scattering collimator factor. CONCLUSIONS: The amount of leakage was more for 18 MV compare to 6 MV The percentage of the leakage for both energies is less than the 5% value which is recommended by protocols. The fluctuation of the MLC leakage reduced by increasing the field size and depth.

Assessment of gamma-dose rate in city of Kermanshah.

INTRODUCTION: Environmental natural radiation measurement is of great importance and interest especially for human health. The induction of genetic disorder and cancer appears to be the most important in an exposed population. MATERIALS AND METHODS: Measurements of background gamma rays were performed using a mini-rad environmental survey meter at 25 different locations around the city of Kermanshah (a city in the west of Iran). The measurements were also performed at two different time of day one in the morning and the other in the afternoon. At each location and time measurements were repeated for five times and the mean was considered as the background dose at that location. RESULTS AND DISCUSSIONS: Comparison between the measured results in the morning and afternoon has not shown any significant difference (P > 0.95). The maximum and minimum obtained results were 2.63 mSv/y and 1.49 mSv/y, respectively. From the total measurements at 25 sites mean and SD background radiation dose to the population is 2.24 ± 0.25 mSv. CONCLUSION: The mean radiation dose to the population is about 2.5 times of the world average total external exposure cosmic rays and terrestrial gamma rays dose reported by UNSCEAR.

Investigation of the bnp level changes in blood stream in different modes and lead locations after pacemaker implementation.

As has been proven, increase of mechanical strain could result in an increase of brain natriuretic peptide (BNP) in the blood stream of implanted patient pacemakers. We measured the BNP concentration in blood due to different mode and lead implantation location of pacemaker in the time period of 3 months. The aim of this study was to investigate the changing trend of BNP level after pacemaker implantation. One hundred and three pacemaker implanted patients were monitored. Patients were in the age span of 54±12 years, including 48 men and 55 women. A group of 44 were programmed in Dual Chamber Rate Adaptive (DDDR) Pacemaker mode and a group of 59 were programmed in Ventricular Rate Modulated Pacing (VVIR) mode by the recommendation of the cardiologist. Between these two groups, the pacing levels of pacemakers was divided to under and above 50%. Some of these pacemaker leads were located at the apex of the right ventricle and the others were located in the septum wall in the right ventricle. To evaluate BNP changes during a period of 3 months, the BNPs were measured in pg/ml within 24 h of implantation (BNP1) and after 3 months (BNP2). For different classes of pacemaker implantations, the ratio of final measurement (BNP2) is divided to after implantation measurements (BNP1). Results showed that in VVIR mode, the ratio is 1.54±0.3 and in DDDR mode, the ratio is 0.38±0.17, with acceptable standard error means (<0.04). Also, comparisons are made for lead location at two modes of DDDR and VVIR separately. In the DDDR mode, the ratio for apex location is 0.49±0.12 and for septum location is 0.22±0.34, with acceptable standard error means (<0.02). In the VVIR mode, the ratio for apex location is 1.71±0.27 and for septum location is 1.28±0.09, with acceptable standard error means (<0.04). Therefore, BNP decrease in DDDR mode is more than in VVIR mode programming. In both cases of DDDR and VVIR modes, the septum location of the leads would result in a greater decrease of BNP.