Deformable registration of magnetic resonance images using unsupervised deep learning in neuro-/radiation oncology.

PURPOSE: Accurate deformable registration of magnetic resonance imaging (MRI) scans containing pathologies is challenging due to changes in tissue appearance. In this paper, we developed a novel automated three-dimensional (3D) convolutional U-Net based deformable image registration (ConvUNet-DIR) method using unsupervised learning to establish correspondence between baseline pre-operative and follow-up MRI scans of patients with brain glioma. METHODS: This study involved multi-parametric brain MRI scans (T1, T1-contrast enhanced, T2, FLAIR) acquired at pre-operative and follow-up time for 160 patients diagnosed with glioma, representing the BraTS-Reg 2022 challenge dataset. ConvUNet-DIR, a deep learning-based deformable registration workflow using 3D U-Net style architecture as a core, was developed to establish correspondence between the MRI scans. The workflow consists of three components: (1) the U-Net learns features from pairs of MRI scans and estimates a mapping between them, (2) the grid generator computes the sampling grid based on the derived transformation parameters, and (3) the spatial transformation layer generates a warped image by applying the sampling operation using interpolation. A similarity measure was used as a loss function for the network with a regularization parameter limiting the deformation. The model was trained via unsupervised learning using pairs of MRI scans on a training data set (n = 102) and validated on a validation data set (n = 26) to assess its generalizability. Its performance was evaluated on a test set (n = 32) by computing the Dice score and structural similarity index (SSIM) quantitative metrics. The model's performance also was compared with the baseline state-of-the-art VoxelMorph (VM1 and VM2) learning-based algorithms. RESULTS: The ConvUNet-DIR model showed promising competency in performing accurate 3D deformable registration. It achieved a mean Dice score of 0.975 ± 0.003 and SSIM of 0.908 ± 0.011 on the test set (n = 32). Experimental results also demonstrated that ConvUNet-DIR outperformed the VoxelMorph algorithms concerning Dice (VM1: 0.969 ± 0.006 and VM2: 0.957 ± 0.008) and SSIM (VM1: 0.893 ± 0.012 and VM2: 0.857 ± 0.017) metrics. The time required to perform a registration for a pair of MRI scans is about 1 s on the CPU. CONCLUSIONS: The developed deep learning-based model can perform an end-to-end deformable registration of a pair of 3D MRI scans for glioma patients without human intervention. The model could provide accurate, efficient, and robust deformable registration without needing pre-alignment and labeling. It outperformed the state-of-the-art VoxelMorph learning-based deformable registration algorithms and other supervised/unsupervised deep learning-based methods reported in the literature.

Qualitative and Quantitative Investigation of Biochar-Cu0 Composite for Nickel Adsorption.

The current investigation deals with the treatment of water pollution that is caused by the leaching of nickel ions from the metallurgical industry and new-energy batteries. Therefore, an eco-friendly treatment of nickel through the use of a composite of cotton stalk biochar with nanozerovalent copper has been presented in this investigation signifying the impact of zerovalent copper in enhancing the adsorption capacity of biochar for nickel adsorption. Thermogravimetric analysis data showed the adsorbent to be significantly stable in the higher thermal range, whereas transmission electron microscopy analysis confirmed the particles to be 27 nm and also showed the cubic geometry of the particles. A much closer scanning electron microscopy analysis shows the morphology of particles to be cubic in shape. Batch adsorption indicated a positive influence of pH increase on adsorption due to the electrostatic attraction between positive nickel ions and post point of zero charge (pHPZC) negative surface of copper biochar composite (pH > 5.5). A high adsorption rate was observed in the first 60 min, whereas adsorption increased with the increase in temperature from 303 to 318 K. Kinetic modeling confirmed the pseudo-first-order to fit best to the data. The apparent activation energy (11.96 kJ mol-1) is indicative of the chemical nature of the process. The adsorption data fitted well to the Langmuir adsorption model. The negative values of apparent ΔG° and the positive values of apparent ΔH° indicate the spontaneity and endothermicity of the process, respectively, whereas the positive values of apparent ΔS° point toward increased randomness during the process. Postadsorption XPS suggests the adsorption of nickel on the surface of biochar composites in the form of Ni(OH)2 and NiO(OH).

Natural polymers as potential P-glycoprotein inhibitors: Pre-ADMET profile and computational analysis as a proof of concept to fight multidrug resistance in cancer.

P-glycoprotein (P-gp) is known as the "multidrug resistance protein" because it contributes to tumor resistance to several different classes of anticancer drugs. The effectiveness of such polymers in treating cancer and delivering drugs has been shown in a wide range of in vitro and in vivo experiments. The primary objective of the present study was to investigate the inhibitory effects of several naturally occurring polymers on P-gp efflux, as it is known that P-gp inhibition can impede the elimination of medications. The objective of our study is to identify polymers that possess the potential to inhibit P-gp, a protein involved in drug resistance, with the aim of enhancing the effectiveness of anticancer drug formulations. The ADMET profile of all the selected polymers (Agarose, Alginate, Carrageenan, Cyclodextrin, Dextran, Hyaluronic acid, and Polysialic acid) has been studied, and binding affinities were investigated through a computational approach using the recently released crystal structure of P-gp with PDB ID: 7O9W. The advanced computational study was also done with the help of molecular dynamics simulation. The aim of the present study is to overcome MDR resulting from the activity of P-gp by using such polymers that can inhibit P-gp when used in formulations. The docking scores of native ligand, Agarose, Alginate, Carrageenan, Chitosan, Cyclodextrin, Dextran, Hyaluronic acid, and Polysialic acid were found to be -10.7, -8.5, -6.6, -8.7, -8.6, -24.5, -6.7, -8.3, and -7.9, respectively. It was observed that, Cyclodextrin possess multiple properties in drug delivery science and here also demonstrated excellent binding affinity. We propose that drug efflux-related MDR may be prevented by the use of Agarose, Carregeenan, Chitosan, Cyclodextrin, Hyaluronic acid, and/or Polysialic acid in the administration of anticancer drugs.

Development of a Novel Design and Subsequent Fabrication of an Automated Touchless Hand Sanitizer Dispenser to Reduce the Spread of Contagious Diseases.

BACKGROUND: The use of a touchless automated hand sanitizer dispenser may play a key role to reduce contagious diseases. The key problem of the conventional ultrasonic and infra-red-based dispensers is their malfunctioning due to the interference of sunlight, vehicle sound, etc. when deployed in busy public places. To overcome such limitations, this study introduced a laser-based sensing device to dispense sanitizer in an automated touchless process. METHOD: The dispensing system is based on an Arduino circuit breadboard where an ATmega328p microcontroller was pre-installed. To sense the proximity, a light-dependent resistor (LDR) is used where the laser light is to be blocked after the placement of human hands, hence produced a sharp decrease in the LDR sensor value. Once the LDR sensor value exceeds the lower threshold, the pump is actuated by the microcontroller, and the sanitizer dispenses through the nozzle. RESULTS AND DISCUSSION: A novel design and subsequent fabrication of a low-cost, touchless, automated sanitizer dispenser to be used in public places, was demonstrated. The overall performance of the manufactured device was analyzed based on the cost and power consumption, and environmental factors by deploying it in busy public places as well as in indoor environment in major cities in Bangladesh, and found to be more efficient and cost-effective compared to other dispensers available in the market. A comprehensive discussion on this unique design compared to the conventional ultrasonic and infra-red based dispensers, is presented to show its suitability over the commercial ones. The guidelines of the World Health Organization are followed for the preparation of sanitizer liquid. A clear demonstration of the circuitry connections is presented herein, which facilitates the interested individual to manufacture a cost-effective dispenser device in a relatively short time and use it accordingly. Conclusion: This study reveals that the LDR-based automated hand sanitizer dispenser system is a novel concept, and it is cost-effective compared to the conventional ones. The presented device is expected to play a key role in contactless hand disinfection in public places, and reduce the spread of infectious diseases in society.