Medical Image Processing

In this Paper, we present a new approach to protect digital images using the genetic algorithm using the image encryption technique. In genetic algorithms, the crossing operation assembles the existing genes and the mutation operation produces new genes. Here, for each pair of pixels of digital image, the binary value of a cross-image is applied and the mutation is used to encrypt the digital image. Several test analyzes are performed, such as correlation analysis, entropy, and histogram, to verify the performance of the digital protection of the medical images or the encryption algorithm.

Penelitian ini bertujuan untuk meningkatkan kinerja model CNN (jaringan lipat) dalam tugas klasifikasi gambar menggunakan modul perhatian, terutama modul catatan blok lipat (CBAM). Model ini diuji menggunakan dataset CIFAR 10 yang terdiri dari 60.000 gambar warna dengan 10 kelas objek yang berbeda. Kami membandingkan dua arsitektur: CBAM-ditingkatkan standar-CNN dan CNN. Hasil pelatihan menunjukkan bahwa integrasi CBAM dapat meningkatkan akurasi tes dari sekitar 74% menjadi 80,78%, yang meningkatkan kompromi dalam bentuk sedikit peningkatan inferensi (~ 23%) per ERA. Visualisasi kartu catatan juga menunjukkan bahwa model yang menggunakan CBAM dapat fokus pada fitur -fitur utama foto, seperti garis besar tujuan utama. Temuan ini menunjukkan bahwa penerapan modul perhatian ke CNN memberikan peningkatan kinerja yang signifikan dalam klasifikasi gambar dan dapat diterapkan pada berbagai aplikasi nyata, seperti pengenalan objek otomatis dan sistem pengawasan visual. Namun, dalam aplikasi nyata, peningkatan kompleksitas dan penundaan harus dipertimbangkan.

Tomographic phase microscopy measures the 3-D refractive index distribution of cells and tissues by combining the information from a series of angle-dependent interferometric phase images. In the original device, the frame rate was limited to 0.1 frames per second (fps) by the technique used to acquire phase images, preventing measurements of moving or rapidly changing samples. We describe an improved tomographic phase microscope in which phase images are acquired via a spatial fringe pattern demodulation method, enabling a full tomogram acquisition rate of 30 fps. In addition, in this system the refractive index is calculated by a diffraction tomography algorithm that accounts for the effects of diffraction in the 3-D reconstruction. We use the instrument to quantitatively monitor rapid changes in refractive index within defined subregions of cells due to exposure to acetic acid or changes in medium osmolarity. C 2011 Society of Photo-Optical Instrumentation Engineers (SPIE).

Purpose: Many methods are available to segment structural magnetic resonance (MR) images of the brain into different tissue types. These have generally been developed for research purposes but there is some clinical use in the diagnosis of neurodegenerative diseases such as dementia. The potential exists for computed tomography (CT) segmentation to be used in place of MRI segmentation, but this will require a method to verify the accuracy of CT processing, particularly if algorithms developed for MR are used, as MR has notably greater tissue contrast. Methods: To investigate these issues we have created a three-dimensional (3D) printed brain with realistic Hounsfield unit (HU) values based on tissue maps segmented directly from an individual T1 MRI scan of a normal subject. Several T1 MRI scans of normal subjects from the ADNI database were segmented using SPM12 and used to create stereolithography files of different tissues for 3D printing. The attenuation properties of several material blends were investigated, and three suitable formulations were used to print an object expected to have realistic geometry and attenuation properties. A skull was simulated by coating the object with plaster of Paris impregnated bandages. Using two CT scanners, the realism of the phantom was assessed by the measurement of HU values, SPM12 segmentation and comparison with the source data used to create the phantom. Results: Realistic relative HU values were measured although a subtraction of 60 was required to obtain equivalence with the expected values (gray matter 32.9-35.8 phantom, 29.9-34.2 literature). Segmentation of images acquired at different kVps/mAs showed excellent agreement with the source data (Dice Similarity Coefficient 0.79 for gray matter). The performance of two scanners with two segmentation methods was compared, with the scanners found to have similar performance and with one segmentation method clearly superior to the other. The ability to use 3D printing to create a realistic (in terms of geometry and attenuation properties) head phantom has been demonstrated and used in an initial assessment of CT segmentation accuracy using freely available software developed for MRI.

Although this huge development in medical imaging tools, we find that there are some human mistakes in the process of filming medical images, where some errors result in distortions in the image and change some medical image properties which affect thedisease diagnosis correctly.Medical images are one of the fundamental images, because they are used in the most sensitive field which is a medical field. The objective of the study is to identify the effect of implement non-linear filters in enhancingmedical images, using the strongest and most popular program MATLAB, and because of its advantages in image processing. After implementation the researcher concluded that we will get the best result for medical image enhancement by using median filter which is one of the non-linear filters,non-linear filters implemented using Matlab functions.

In order to begin the initial check on breast cancer, radiologist can use Computer Aided Diagnosis (CAD) as another option to detect breast cancer. During breast cancer check, human error is often to affecting the result. Several research before have proved that CAD is able to detect breast cancer spot more accurate. The purpose of this research is to find reliable method to classify breast cancer abnormalities. Mammography Image Analysis Society (MIAS) database is used as the sample data to the proposed system in this research. Mammograms are divided into three categorize which are normal, benign and malignant according to MIAS database. Features included in this experiment are extracted by using gray level co-occurrence matrices (GLCM) at 0o, 45o, 90o and 135o with a block size of 128x128. In classification process, this research attempt to compare k-Nearest Neighbor (k-NN) and Support Vector Machine (SVM) classifier in order to achieve the better accuracy. The result shows that S...

Te emergence of deep learning has markedly enhanced the identifcation and diagnosis of ocular diseases, providing considerable benefts compared to conventional machine learning techniques. Tis research investigates the application of deep feature extraction for classifying eight diferent ocular diseases. Te VGG16, a pretrained convolutional neural network (CNN) model, was employed for feature extraction, while the fne k-nearest neighbor (KNN) classifer was utilized for classifcation. Experimental results showed an initial classifcation accuracy of 89.2% using features from Gradient-weighted Class Activation Mapping (Grad-CAM) heatmaps and 83.1% using Multiscale Retinex (MSR) enhanced images. However, combining both feature sets led to an improved classifcation accuracy of 96.5%. Despite these promising results, several challenges remain, including the need for models that generalize across diverse patient imaging and demographics modalities, the accessibility of extensive annotated datasets, and the interpretability of models. Ethical issues and legal frameworks are also crucial for the safe and fair implementation of AI in medical services. Te study suggests that future eforts in deep learning for ophthalmology should focus on creating large-scale, annotated datasets to enhance the detection of ocular diseases.

The outcome of simulated experiments expressed the efficiency of grid-based algorithm in the domain of image processing in the context of Brain MRI Images for Tumor Detection. It provides a faster approach in detecting object and achieving higher rate of detection in comparison with existing algorithms for the similar purpose .The foremost goal of this work is improving the rate of detection by using grid based feature fusion approach with Genetic Algorithm(GA), Particle Swarm Optimization(PSO) and Scalp Swarm Optimization Algorithm. This technique produces efficient result due to hybrid fusion of gridded Images and optimization for exhaustive and precise image analysis. In the initial phase Each image is segmented into 4 grids. Pretrained models like Inception-V2, Resnet 50, Mobnetv2, VGG-16 are used to extract features before and after the images are gridded. The extracted features are analysed with classifiers such as Decision tree, Support vector Machine, Multiplayer and Self Attention Transformer that not only chooses the best set of features as well as best feature extractor. Among all these models the VGG-16 gives best result in terms of Accuracy, sensitivity, specificity and F-score. Hence VGG-16 is utilized as evaluation functions in the optimization Algorithms. The performance of the approach is measured by the values produced against the terms as precision, Recall, F-score, and accuracy rate. In the first phase of iteration, features extracted from each grid are assigned with initial weight. Optimization algorithms are used for optimizing the weight values in the combined features to bring best accuracy ratio. Grid based segmentation of Image stabilize the computation time thereby reduces the time complexity. Weighted sum Scalp Swarm Optimisation assigns best weighted values to segmentation parameters to get maximum performance. The experimental outcomes proves the efficiency of the proposed strategy .

Since the discovery of the microscope, analysis towards cell biology is one of the integral parts for the understanding of cell behavior. These analyses of the cells biology have two main perspectives comprising of quantitative and qualitative factors. Cell segmentation is one of the critical tasks which make ease the process of cell analysis. Phase-contrast Microscopy(PCM)  images have an integral role in a detailed analysis of the cell structures as it allows a wide range of possibilities in studying living cells from different perspectives in a label-free manner. Segmentation of the PCM images has been challenging for the biologists due to problems specific to the examined specimen or the microscopy technique used,  such as perturbations in the cell shapes,  overlapping and transparent appearance of the cells in PCM images which complicates the cell segmentation with conventional computer vision approaches, effective extraction of the Region of Interest (RoI) from images. Therefore, most of the segmentation and tracking solutions available in the literature do not qualify as satisfactory in terms of robustness, completeness, and accuracy.

The recently popular Deep Neural Networks(DNNs) have a significant effect on the improvement of segmentation accuracy from different perspectives, including robustness and completeness compared to conventional methods; moreover, deep learning-based approaches are also unbeaten in classification, detection and tracking applications. However, deep learning bases approaches require large numbers of annotated datasets to converge at high performances. By the way, applying the specific architectures which are well-known in the field of medical imaging segmentation can be successful even in low annotated images efficiently.

We proposed an alternative feature extractor by applying the modified ResNet18 in the encoder of U-Net and replacing the plain blocks with residual blocks in the decoder of the U-Net architecture. As an extension of the ResNet18-U-Net, we employed a residual pathway as an alternative to the skip connection of the model, which made the model effective in taking low and high-level semantics and the detail of the images into account. Furthermore, we utilized ResNet18 in the backbone of the Feature Pyramid Network(FPN) which has higher performance in comparison to conventional methods.



In our experimental results, we employ our proposed models in comparison to the base-line models on MDA-MB-231 and DSB2018 dataset. Our proposed model achieved Jaccard Index of 85.4\% $\pm$ 2.7, 87.9\% $\pm$  1.7, 87.1\% $\pm$  2.3 and 89.2\% $\pm$ 1.3 with TipNet, ResNet18-U-Net, ResNet18-FPN and ResNet18-U-Net+RP  respectively on MDA-MB231 dataset and achieve the Jaccard index of 83.1\% $\pm$ 1.8, 81.2\% $\pm$ 2.2 and 85\%$\pm$ 2.1 in the DSB2018 dataset which is slightly less than BCDUNet in the Jaccard Index metric. All of the determined results are come out after applying the TTA as post processing at the proposed method on the DSB2018 and MDA-MB231 datasets. However, our proposed method passes the BCDUNet on MDA-MB-231 dataset.  Moreover, the proposed approach is provided satisfactory segmentation results in terms of robustness and completeness on the MDA-MB231 dataset and DSB2018 dataset. Moreover, our proposed methods are also robust in the segmentation of the boundary regions of the cells in the presented dataset; from the other perspective, these methods are reliable on the detection of the outlier and it causes to reduce the amounts of false prediction in comparison of baseline methods. Among the proposed methods ResNet18-U-Net+RP is outperforming in the separation of the boundaries with faint characteristics.

Weed is one of the main reason for getting less production in agriculture field. At present, farmers are using herbicides to control the weed but it's having negative impact on crop production. To increase the crop production, farmers want to reduce the usage of herbicides. One of the machine learning algorithm, convolution neural network used to classify the weed and crop with high accuracy. CNN gives the best performance in some critical situations also, such as collected images of different lighting conditions, identification of plant species, overlapped crop with weed and designing an autonomous patch sprayers. This review article presents a brief overview of some significant research efforts in weed detection system by using image processing techniques and convolution neural network in machine learning algorithms. Not only CNN, this article reviewed different successful machine learning techniques of supervised and unsupervised learning approaches such as support vector machines, random forest and artificial neural network. This paper aims at providing different challenges and imperatives of various convolution neural network algorithms used in weed detection. Finally, compared different convolution neural network architectures and finds the best CNN architecture used for weed identification system with respect to accuracy.

The study seeks to create and utilize sophisticated deep learning model namely Convolutional Neural Network (CNN) designed for image classification purposes. It's a custom CNN architecture tailored for the task, to classify images of blood cells into eight distinct categories. It comprises several layers, starting with convolutional layers followed by max-pooling layers, and fully connected layers empowered by AI technique, aiming to precisely forecast different forms of blood cancer using a dataset comprising images of blood cells. Ensuring the consistency and quality of input data, especially in medical imaging datasets susceptible to variations affecting model performance, poses a significant challenge; improving the interpretability and explainability of deep learning models is crucial for their reliable incorporation into clinical decisionmaking processes for blood cancer prediction. Utilize sophisticated deep learning model design methodologies guided by Artificial Intelligence approaches, tailored to address the unique characteristics of the blood cell image dataset, in order to surmount the outlined challenges. The suggested system provides improved precision and effectiveness in predicting blood cancer by employing advanced deep learning models guided by customized Artificial Intelligence techniques. Throughout the span of 50 epochs, the model's performance was assessed based on its training and validation accuracy, gradually enhancing until reaching final rates of about 90.31% for training and 90% for validation.

The breast cancer constitutes the most frequent cause of woman's death in developing countries, and 35% of women age 35 -55 in industry countries. The best way to reduce the death rate caused by carcinoma breast is by early detection and treatment. Computer assisted diagnosis system gives a great possibility for the equipment needed. Fuzzy Edge Detection is one of many operations done in image processing. It's been used to identify the object in the image which makes it possible to characterize and detect the tumor based on abnormal structure. However this method should not be applied without other application's assistance such as morphological watershed algorithm for its accuracy.

Empfohlene Zitierung / Suggested Citation: Burri, R. (2003). Digitalisieren, disziplinieren: soziotechnische Anatomie und die Konstitution des Körpers in medizinischen Bildgebungsverfahren (TUTS Working Papers, 3-2003). Berlin: Technische Universität Berlin, Fak. VI Planen, Bauen, Umwelt, Institut für Soziologie Fachgebiet Techniksoziologie. https://nbn-resolving.org/ urn:nbn:de:0168-ssoar-11594

Image-guided intervention and needle steering for radiofrequency ablation (RFA) of the liver is reviewed in this paper. In particular, the concept of active needle is proposed for RFA treatment. Methods and techniques of image processing and modeling are presented for a stereo liver model. The liver model and constituent components extracted from computerized tomography (CT) images can be used to plan the navigation paths of the RFA needle. The system also provides an option for active needles, which are more amenable to those refractory cases of RFA treatment.

An approach aimed at detecting types of brain abnormalities in MRI images is presented, which provides high performance and simplicity of design. The developed model proposes to extract handcrafted numerical features from MRI images using a fine-tuned histogram of oriented gradients algorithm. The obtained features undergo dimensionality optimization by applying principal component analysis before being passed to a 6-layer custom neural network that is trained to perform the tumor identification task. In addition, our work investigates the feasibility of performing augmentation in the feature space using the synthetic minority oversampling technique. The proposed approach achieved a high accuracy of 99.65 % and an F1-score of 99.64 %, superior to many works in recent literature.

In the paper we present an approach to Introduce automation of brain CT image analysis Because CT Scan method that used especially for the diagnosis of stroke and can detect bleeding in stroke due to a blocked artery, of course Images from a CT scan resolution is low relatively. Therefore, the grayscale images resolution is scant and makes detection difficult. We can use bioinformatics and artificial coloring techniques by image processing quality added and is more sensitive in outstanding. We have to identify and distinguish the areas of clustering artificial colors with Hopfield clustering that introduced as Pixel clustering based segmentation method and improve it by Hopfield neural network (HNN) based on spectral properties to show different region by artificial coloring and clustering. We want to improve the technique to use this rule by determining best cluster in neural network.

Medical image classification has become a critical task in computeraided diagnosis, enabling faster and more accurate detection of diseases such as cancer, pneumonia, and diabetic retinopathy. This study presents a comprehensive comparative analysis of prominent neural network architectures-Convolutional Neural Networks (CNNs), Residual Networks (ResNets), DenseNets, Vision Transformers (ViTs), and EfficientNets-in the context of medical image classification. Utilizing benchmark datasets including ChestX-ray14, ISIC Skin Cancer, and Retinal Fundus Images, we evaluated each model's performance based on accuracy, precision, recall, F1-score, training efficiency, and robustness to overfitting. The results demonstrate that while CNN-based models like ResNet and DenseNet maintain strong classification capabilities with balanced computation cost, ViTs outperform others in high-resolution image interpretation, especially under complex feature distributions. EfficientNet offers a trade-off between speed and accuracy, making it suitable for resource-constrained clinical settings. Our findings highlight the architectural strengths and weaknesses in varied medical imaging scenarios, providing insights into the selection of optimal models based on diagnostic goals, dataset characteristics, and computational resources..

The Medical image database is growing day by day [1]. Most of the medical images are stored in DICOM (Digital Imaging and Communications in Medicine) format. There are various categories of medical images such as CT scan, XRay, Ultrasound, Pathology, MRI, Microscopy, etc [2]. Physicians compare previous and current medical images associated patients to provide right treatment. Medical Imaging is a leading role in modern diagnosis. Efficient image retrieval tools are needed to retrieve the intended images from large growing medical image databases. Such tools must provide more precise retrieval results with less computational complexity. This paper compares the various techniques for DICOM medical image retrieval and shows that the proposed geometric and image segmentation based image retrieval algorithm performs better as compared to other algorithms.

Nous revisitons les techniques de tatouage numérique basées sur des transformations d'images, à la lumière des réseaux neuronaux auto-supervisées. Nous présentons un moyen d'intégrer à la fois des marques et des messages binaires dans leurs espaces latents. Notre méthode peut fonctionner à résolution variable et crée des tatouages robustes à un large éventail de transformations (rotations, recadrages, JPEG, contraste, etc.). Elle surpasse de manière significative les précédentes méthodes de tatouage zéro-bit, et ses performances en multi-bits sont comparables à celles des architectures encodeur-décodeur entraînées de bout en bout pour le tatouage numérique. Le code est disponible sur github.com/facebookresearch/ssl_watermarking.

In numerous therapeutic explores and applications, division and arrangement of brain tissue in MRI is a troublesome undertaking 4 . Besides, manual division and order of MRI brain images is a lengthy procedure. MRI cerebrum images have their applications in the fields such

Adders play a pivotal role in numerous digital and high-speed systems, including microprocessors, digital signal processors, and FIR filters. They yield substantial influence over the speed, footprint, and energy usage of digital systems. Striking a balance between speed, size, power consumption, and other variables is imperative given the multifaceted demands placed on digital systems. The procedure is producing partial products, lowering these products, and then carrying out carry-propagate addition. When designing adders, the goal is to minimize power dissipation without sacrificing other performance parameters. Reducing dynamic power dissipation is crucial for achieving significant power reductions in VLSI design. The goal of this paper is to choose a suitable adder that is affordable, fast, and uses less power. In this paper the performance of two 16-bit adders; such as Ripple Carry Adder and Carry Select Adder are compared based on on-chip power, latency in datapath and the number of look up tables (LUT)s used by each adder. The design and stimulus blocks of the adders are written in Verilog HDL and simulated in Vivado 2023.1 tool. Furthermore, the adders are implemented on AMD Zynq 7000 SoC ZC702 Evaluation Kit FPGA board. Upon execution of the formerly said procedure, it is found that the Ripple carry adder outperforms carry select adder in terms of delay and the LUTs used. On the other hand, Ripple carry adder has 144.9W on-chip power consumption while the carry select adder consumes an on-chip power of 132.42W.

Circular data, i.e., data in the form of 'natural' directions or angles are very common in a number of dierent areas such as biological, meteorological, geological, and political sciences. Clustering circular data is not an easy task due to the circular geometry of the data space. Some clustering approaches, such as the spherical k-means, use the cosine distance instead of the euclidean distance in order to measure the dierence between points. In this paper, we propose a variation of the randomized gravitational clustering algorithm in order to deal with circular data. Basically, we use the cosine distance, we modify the gravitational law in order to use the cosine distance and we use geodesics ('straight' lines in curved spaces) in order to move points according to the gravitational dynamic. Our initial experiments indicate that the spherical gravitational clustering algorithm is able to nd clusters in noisy circular data.

Financial fraud has emerged as a major challenge in today's digital economy, with an increasing number of fraudulent activities targeting online financial systems. This study proposes a unified deep learning approach for detecting fraudulent financial transactions using advanced neural network architectures. Traditional fraud detection methods rely heavily on rule-based or shallow learning algorithms, which often fail to detect novel fraud patterns. In contrast, this research introduces a hybrid framework incorporating convolutional neural networks (CNNs), gated recurrent units (GRUs), and attention mechanisms to capture both spatial and temporal dependencies within transaction sequences. We use a benchmark dataset of anonymized financial transactions, apply comprehensive preprocessing steps including normalization, class balancing, and feature engineering, and evaluate model performance using multiple metrics: RMSE, MAPE, and R^2. Experimental results show that the unified model outperforms conventional machine learning techniques and individual deep learning models in terms of accuracy and robustness. Furthermore, visualizations such as confusion matrices, ROC curves, and prediction plots are used to interpret model effectiveness. This work demonstrates that a unified deep learning strategy not only enhances detection performance but also provides a scalable solution for real-world financial institutions. Our findings highlight the necessity of integrating multiple deep learning architectures to address complex fraud scenarios effectively. Future work aims to extend this model to multimodal data sources such as social behavior and geolocation for enhanced fraud profiling.

Medical imaging provides physicians with the ability to generate 3D images of the human body in order to detect and diagnose a wide variety of ailments. Making medical imaging portable and more accessible provides a unique set of challenges. In order to increase portability, the power consumed in image acquisitioncurrently the most power-consuming activity in an imaging device -must be dramatically reduced. This can only be done, however, by using complex image reconstruction algorithms to correct artifacts introduced by low-power acquisition, resulting in image processing becoming the dominant power-consuming task. Current solutions use combinations of digital signal processors, generalpurpose processors and, more recently, general-purpose graphics processing units for medical image processing. These solutions fall short for various reasons including high power consumption and an inability to execute the next generation of image reconstruction algorithms. This paper presents the MEDICS architecture -a domain-specific multicore architecture designed specifically for medical imaging applications, but with sufficient generality to make it programmable. The goal is to achieve 100 GFLOPs of performance while consuming orders of magnitude less power than the existing solutions. MEDICS has a throughput of 128 GFLOPs while consuming as little as 1.6W of power on advanced CT reconstruction applications. This represents up to a 20X increase in computation efficiency over current designs.

This article presents the design and implementation of a system for detection of license plates for a public parking located at the municipality of Altamira at the state of Huila in Colombia. The system includes also, a module of surveillance cameras for remote monitoring. The detection system consists of three steps: the first step consists in locating and cutting the vehicle license plate from an image taken by an IP camera. The second step is an optical character recognition (OCR), responsible for identifying alphanumeric characters included in the license plate obtained in the first step. And the third step consists in the date and time register operation for the vehicles entry and exit. This data base stores also the license plate number, information about the contract between parking and vehicle owner and billing.

Iron deficiency is one of the most important problems that affects the growth of plants. In this study, the effect of two strains of growth stimulating bacteria producing siderophore (Enterobacter cloacae & Bacillus cereus) on biochemical properties of pistachio seedlings in conditions of availability and deficiency of iron was investigated.Methods: In this research, which was done in a factorial design in the form of a completely randomized pattern, ferric iron (Fe + 3) and ferrous iron (Fe + 2) were given to plants that had grown for two months at a concentration of 20 micromolar. Plants without iron (zero iron) were considered as controls.Results: The results showed that ferric iron in concentrations of 20μM and zero iron reduced plant growth and damage to them. By using growth-promoting bacteria in ferric iron-treated plants, some growth parameters such as fresh weight, photosynthetic pigments, phenolic compounds, soluble sugars, lipid peroxidation and proline were improved. Treatment of plants with growth-promoting bacteria also increased the amount of protein in plants treated with ferric iron. Growth-promoting bacteria also increased iron uptake in the shoots of ferric iron-treated seedlings. It seems that in plants treated with ferric iron, growth-promoting bacteria with the production of siderophore led to the reduction of iron and increase the available iron in the plant and reduce iron deficiency stress.

- The atomic number of an atom is the number of protons in the nucleus  Isotopes of an element have the same atomic number but different mass number  The SI unit of radioactivity is the becquerel (Bq). 1 Bq = 1 transformation/second. Practical units are kBq, MBq, GBq : , λ = ln2/t1/2 = 0.693/t1/2 , A = λN , A(t) = Aoe-λt  In radioactive decay the parent radionuclide is transformed into a new nuclide. The daughter may be stable or itself radioactive

- - Aim: The concept of atoms as the ultimate smallest, not further divisible constituent of matter is introduced. It arises from more than two millennia old Greek philosophy, reaches the atom’s renaissance in the 19th century, and reflects the dramatic improvements achieved at the beginning of the 20th century.

- Atoms are made of electrons, which exist in the shell of an atom, and a set of different particles located in the atom’s nucleus. To understand the chemical properties of an atom, the electron shell structure, i.e. the number and the characteristics of electrons in various shells and the transition of electrons between shells, is essential. To a large extent, this understanding needs a significant
reflection on quantum physics.

- Developments of concepts of ancient and modern atom theory are introduced, turning from the atomic philosophy of being the ultimate particle to concepts on the atom being a substance of various, subatomic particles. The latter ideas are illustrated as a scientific need to understand pioneering experiments in chemistry and physics, i.e. to correlate experimental evidence and subsequent theoretical explanation.

- The latter is meant too to serve as an introduction to quite similar considerations relevant to the structure of nucleons in the atomic nucleus.

- Unstable nuclei convert into stable ones by minimizing the absolute mass of a nuclide, which is typically expressed in terms of nucleon binding energy. The common feature of all these transformations is their exothermic character, and transformations occur spontaneously. In contrast, nuclear reactions start through external “activation”, delivered by a projectile. Balances in mass of reactants and products may differ between exothermic and endothermic routes, and in the latter case the projectile delivers activation in the form of kinetic energy.

- The general strategy of artificial nuclear transformation is to induce a nuclear reaction by delivering energy (either in the form of a particle or electromagnetic) of adequate kinetic energy into a (mainly) stable target atom. Thereby, a compound nucleus is formed, which instantaneously transforms into the nuclear reaction product(s).

- There are several pathways for nuclear reactions and several classes may be defined. Some occur naturally in the universe, in particular in stars or in interstellar space, some in the earth’s atmosphere. Another group subsumes artificial, man-made nuclear reactions.

We discuss the potentially misleading effect of software techniques for elaborating low-contrast images. In particular, we present the example of the stains embedded into one of the most studied archaeological objects in history, the Shroud of Turin. We show for the first time that image processing of both old and recent photographs of the Shroud may lead some researchers to perceive inscriptions and patterns that do not actually exist, confirming that there is a narrow boundary between image enhancement and manipulation.

Images have found widespread use in today’s world from social networking, banking, medical to military uses. Hence, to maintain confidentiality of these images, image encryption becomes an integral part of any system. This paper reviews some of the image encryption techniques and compares the performance of these techniques with respect to different parameters.

Respiratory diseases are one of the most common diseases among population which can affect person's respiratory system. Artificial intelligence (AI) is an innovative technology in medical field which is the realization of complex issue that require information processing skills, such as decision making through data. AI is being used for all the three classical medical tasks that are Diagnosis, Prognosis, Therapy in the area of medical diagnosis. Imaging The use of AI to compound imaging technology has set up success in several disciplines, including computer-backed discovery and opinion (CAD), convolutional neural networks (CNNs), and radiomics. CAD systems are generally standalone with a unified thing of discovery or opinion of complaint. In addition to towel biomarker analysis, liquid necropsies have come more common. Liquid vivisection is analysis of any excrescence deduced product in the blood or serum. For instance, a study using CNNs achieved an accuracy of 89.45% and an AUC of 0.96. Another study analysing LDCT scans from NLST achieved an AUC of 0.94, outperforming radiologists in cases where prior CT imaging was unavailable, leading to reductions in false positives and false negatives.

Breast cancer is one of the leading causes of cancer death in women around the world. In order to reduce the death rate, the tumors have to be detected at the early stage. CAD system are considered to be the best technique for screening. This paper presents a survey of various staging techniques used for early and accurate detection of tumor, which aids in early medication by reducing false positive and false negative discrepancies. The several stages involved in the detection of tumor are pre-processing, segmentation, feature extraction and classification lastly staging of cancer, if confirmed for malignancy. Pre-processing stage improves the quality of the image by removing artifacts, and enhancing the image. Segmentation deals with removal of the pectoral muscle and segmenting the tumor part. Classification is used to classify the detected tumor as Malignant or Benign. The tumors identified as Malignant are then processed using various parameters like size, shape, energy levels and several other features for staging.

Objective: The objective of this research is to automate the calculation of differential blood count in blood smear photomicrographsusing image quantization.Material and Methods: A series of image processing steps were used for the detection of White Blood Cells (WBCs),Red Blood Cells (RBCs) and platelets as: image acquisition, separating the channels of RGB and applying wiener filteron each channel for smoothing the image. The purpose is to enhance the visual interpretation of the image, recombiningthe channels and applying the quantization over the wiener output.Results: The accuracy of this technique is very close to that of the hematologists’ manual calculation. It was 85% forRed blood cells and 98% for White blood cells.Conclusion: This proposed technique gives precise results under varied luminance conditions such as darkness,brightness and low contrast images, it gives reliable results for all the images in the image sets having different qualityof images

Non-proliferative diabetic retinopathy (NDPR) detection is currently a highly interested research area. Ophthalmologists detect NDPR by observing disorders in the vessel system. Therefore segmentation of the vessel system will be an aid for ophthalmologist in order to detect an early retinopathy. In this proposed frame work a novel method based on Gabo Filter and adaptive thresholding has been used. The results have been tested using sensitivity and specificity and the values are 92.36% and 87.52% respectively.

This paper addresses issues of brain tumor subtype classification using Magnetic Resonance Images (MRIs) from different scanner modalities like T1 weighted, T1 weighted with contrast-enhanced, T2 weighted and FLAIR images. Currently most available glioma datasets are relatively moderate in size, and often accompanied with incomplete MRIs in different modalities. To tackle the commonly encountered problems of insufficiently large brain tumor datasets and incomplete modality of image for deep learning, we propose to add augmented brain MR images to enlarge the training dataset by employing a pairwise Generative Adversarial Network (GAN) model. The pairwise GAN is able to generate synthetic MRIs across different modalities. To achieve the patient-level diagnostic result, we propose a post-processing strategy to combine the slice-level glioma subtype classification results by majority voting. A two-stage course-tofine training strategy is proposed to learn the glioma feature using GAN-augmented MRIs followed by real MRIs. To evaluate the effectiveness of the proposed scheme, experiments have been conducted on a brain tumor dataset for classifying glioma molecular subtypes: isocitrate dehydrogenase 1 (IDH1) mutation and IDH1 wild-type. Our results on the dataset have shown good performance (with test accuracy 88.82%). Comparisons with several state-of-the-art methods are also included.

Objective: To evaluate the influence of Body Mass Index on the menstrual pattern in adolescent girls. Methods: Fifty adolescent girls who attained menarche were selected for the study. The data was collected by using a pre-designed pre-tested questionnaire. BMI was calculated using the formula: BMI (kg/m 2 ) = Weight (kg) / Height 2 (m 2 ). Results: The mean age of the girls was 19.96 (range 18 – 21 years). 49% of the fathers worked in the public sector and 2% worked in the private sector, 21% were manual workers and 30% did not mention the status of their father’s job.10% of the adolescent girls had 10 years of age of menarche, 18% of them had 11 years of age of menarche, 42% of them had 12 years of age of menarche, 20% of them had 13 years of age of menarche and 10% of them had 14 years of age at menarche. Nutritional status of the girls was determined by their Body mass index (BMI). 58% of the adolescent girls were normal (having BMI 18.5-24.9), 24% of them were overweight (BMI 2...

The COVID-19 pandemic resulted in one of the most recent economic shocks, impacting global trade, financial markets, and consumer behavior. In the US, GDP suffered a historic downturn in 2020, followed by an unbalanced recovery. Precise GDP growth forecasting has become increasingly essential for policymakers, businesses, and investors making decisions in the postpandemic economy. Classic models, including Autoregressive Integrated Moving Average (ARIMA), Vector Autoregression (VAR), and Dynamic Stochastic General Equilibrium (DSGE), have been popularly employed for GDP forecasting. Machine learning (ML) provides a dominant alternative, with the potential to handle enormous amounts of real-time data, sense nonlinear patterns, and handle economic shocks more effectively than traditional approaches. This paper delves into the potential of ML in GDP forecasting, touching on some key techniques, including neural networks, ensemble learning, and deep learning. This paper assessed the accuracy of two machine learning models, Random Forest (RF) and Long Short-Term Memory (LSTM), in forecasting U.S. GDP growth during the post-COVID-19 pandemic. Although ML-based forecasting holds prominent advantages, challenges, including data quality, explainability, and ethical issues, must be resolved for increased usage in economic decision-making.

This paper introduces a new method for the discrimination of impulse noise in medical MRI images. Statistical Fdistribution by means of Monte Carlo simulation for forecasting the location of Impulse noise on distorted images is employed. The adaptive noisy pixels are identi ed as noisy using One-Way ANOVA hypothesis method while the rest of the noisy pixels are selective noisy pixels. Thus, the statistical prediction method is used to predict the noise location on the test images. The distorted portions in the images predicted by this structure are then made better by applying a right kind of nonlinear ltering. The nonlinear median lter is advisable for demonstrating a suitable way of the replacement of the degraded pixels. The lter is not applied to the not-detected regions of the image to remain the same so that delicate details and edges are not spoiled. Experimental results enhanced this structure as it is compared to the classical noise reduction structures.

Deep learning models, particularly Convolutional Neural Networks (CNNs), have achieved remarkable accuracy in medical image analysis tasks like pneumonia detection from chest X-rays. However, their "black-box" nature and the potential brittleness of common explainability methods (e.g., saliency maps) hinder clinical trust and adoption. This paper proposes and evaluates a methodology for enriching CNNs with mathematically grounded global features derived from Topological Data Analysis (TDA) and Fractal Dimension (FD) analysis, aiming to provide complementary, more robust explanations. We integrate these features, extracted from intermediate layers of a pre-trained ResNet50 fine-tuned for pneumonia detection, with the CNN's own deep features. Our results show that while a simple MLP-based fusion significantly degraded performance (accuracy ~73%), an attention-based fusion mechanism successfully integrated the features, matching the high baseline accuracy (~96%) on the original dataset. The TDA and FD features themselves exhibit statistically significant differences between normal and pneumonia classes (FD p < 5e-7), providing

One of the main challenges in lung cancer radiation therapy is how to reduce the treatment margin but accommodate the geometric uncertainty of moving tumor. 4D-CT is able to provide the full range of motion information for the lung and tumor. However, accurate estimation of lung motion with respect to the respiratory phase is difficult due to various challenges in image registration, e.g., motion artifacts and large interslice thickness in 4D-CT. Meanwhile, the temporal coherence across respiration phases is usually not guaranteed in the conventional registration methods which consider each phase image in 4D-CT independently. To address these challenges, the authors present a unified approach to estimate the respiratory lung motion with two iterative steps. Methods: First, the authors propose a novel spatiotemporal registration algorithm to align all phase images of 4D-CT (in low-resolution) to a high-resolution group-mean image in the common space. The temporal coherence of registration is maintained by a set of temporal fibers that delineate temporal correspondences across different respiratory phases. Second, a super-resolution technique is utilized to build the high-resolution group-mean image with more anatomical details than any individual phase image, thus largely alleviating the registration uncertainty especially in correspondence detection. In particular, the authors use the concept of sparse representation to keep the group-mean image as sharp as possible. Results: The performance of our 4D motion estimation method has been extensively evaluated on both the simulated datasets and real lung 4D-CT datasets. In all experiments, our method achieves more accurate and consistent results in lung motion estimation than all other state-of-the-art approaches under comparison. Conclusions: The authors have proposed a novel spatiotemporal registration method to estimate the lung motion in 4D-CT. Promising results have been obtained, which indicates the high applicability of our method in clinical lung cancer radiation therapy.

The technology world ever things have been improved and modified according to the needs of the user likewise accounting also improves year by years in the current scenario accounting work is simplified with technology. Most advanced technology like Artificial Intelligence, data analytical, cloud accounting software block chains create a change in accounting. Every technology has been a help to improve efficiency, reduce human power, and accurate. Technology makes the world on the internet-based working anywhere and any place.

Visual quality of volume rendering for medical imagery strongly depends on the underlying transfer function. Conventional Windows-Icons-Menus-Pointer interfaces typically refer the user to browse a lengthy catalog of predefined transfer functions or to pain-staking refine the transfer function by clicking and dragging several independent handles. To turn the standard design process less difficult and tedious, this paper proposes novel interactions on a sketch-based interface that supports the design of 1D transfer functions via touch gestures to directly control voxel

Analyzing medical volume datasets requires interactive visualization so that users can extract anatomo-physiological information in real-time. Conventional volume rendering systems rely on 2D input devices, such as mice and keyboards, which are known to hamper 3D analysis as users often struggle to obtain the desired orientation that is only achieved after several attempts. In this paper, we address which 3D analysis tools are better performed with 3D hand cursors operating on a touchless interface comparatively to a 2D input devices running on a conventional WIMP interface. The main goals of this paper are to explore the capabilities of (simple) hand gestures to facilitate sterile manipulation of 3D medical data on a touchless interface, without resorting on wearables, and to evaluate the surgical feasibility of the proposed interface next to senior surgeons (N = 5) and interns (N = 2). To this end, we developed a touchless interface controlled via hand gestures and body postures t...

ÖZET: Günümüzde, artan nüfus ve değişen ihtiyaçlar doğrultusunda firma sayıları giderek artmakta ve firmalar büyümektedir. Bu bağlamda, aynı alanda faaliyet gösteren birçok firma ortaya çıkmakta, bu nedenle firmaların rekabet kabiliyetini artırması gerekmektedir. Bir firma için mevcut müşterinin elde tutulmasına odaklanmak, yeni müşteri kazanmaktan daha maliyetli olmaktadır. Bir müşterinin kaybedilmemesi için en önemli unsurlardan birisi müşteri ilişkileri yönetiminin bir alt dalı olan müşteri şikâyetlerinin iyi bir şekilde yönetilmesinden geçmektedir. Teknolojide meydana gelen gelişmeler doğrultusunda, birçok alanda olduğu gibi müşteri şikâyeti yönetiminde de teknolojiden sıklıkla faydalanılmaktadır ancak bu durum henüz istenilen seviyelere ulaşmamıştır. Bu çalışmada müşteri şikâyeti yönetimi alanına katkı sağlamak için derin öğrenmeden faydalanan özgün modeller geliştirilmiştir. Bu kapsamda, evrişimsel sinir ağı katmanı kullanılarak müşteri yorumlarının hangi şikâyet türünü ilgilendirdiğini tahmin eden bir model geliştirilmiştir. Finans alanındaki bir veri seti kullanılarak analiz edilen modelin hiper-parametreleri Bayesian optimizasyon yöntemi kullanılarak optimize edilmiştir. Farklı derinliklerde geliştirilen modellerle %85.83’lere ulaşan başarı oranı elde edilmiştir. Literatürde benzer veri seti ile yapılan çalışmalar incelendiğinde önerilen modelin, diğer çalışmalara göre üstün olduğu gözlemlenmiştir. ANAHTAR KELİMELER: Müşteri Şikâyet Yönetimi, Derin Öğrenme, Evrişimsel Sinir Ağları, Tüketici Davranışları ///// ABSTRACT: Nowadays, the number of companies is increasing, and companies are growing in line with the increasing population and changing needs. In this context, many companies operating in the same field emerge, thus companies need to enhance their competitive abilities. For a company, focusing on retaining existing customers is more cost-effective than acquiring new customers. One of the most critical elements in not losing a customer is the effective management of customer complaints, which is a sub-branch of customer relationship management. With the advancements in technology, automated systems are frequently used in customer complaint management, as in many areas, but it has not yet reached the desired levels. In this study, novel models using deep learning were developed to contribute to the field of customer complaint management. In this context, a model was created to predict which complaint type customer comments concern, using a convolutional neural network layer. The models were analyzed using a dataset in the field of finance, and the hyper-parameters of the models were optimized using the Bayesian optimization method. Accuracy of up to 85.83% were achieved with models developed at different depths. When compared to studies with similar datasets in the literature, it was observed that the proposed model outperformed other studies. KEYWORDS: Customer Complaint Management, Deep Learning, Convolutional Neural Network, Consumer Behavior

Sugarcane leaf diseases are major agricultural
issues that lead to reduced production and economic losses.
Early detection is crucial for controlling their spread. This
study introduces a solution for identifying sugarcane leaf
diseases using image processing and deep learning techniques.
Preprocessing steps include normalization for pixel scale
stabilization and Gaussian blur for noise reduction. Features
are extracted using a pre-trained DenseNet201 model, and
classification is performed using a Support Vector Machine
(SVM) with an RBF kernel. The proposed hybrid DCNN-SVM
architecture achieves a test accuracy of 96.74%, demonstrating
effectiveness in disease classification. Comprehensive
evaluations highlight the model’s strengths and areas for
improvement, providing a foundation for future research in
plant disease detection.