This paper introduces a hybrid two-stage registration framework for reconstructing three-dimensional (3D) kidney anatomy from macroscopic slices, using CT-derived models as the geometric reference standard. The approach addresses the data-scarcity and high-distortion challenges typical of macroscopic imaging, where fully learning-based registration (e.g., VoxelMorph) often fails to generalize due to limited training diversity and large nonrigid deformations that exceed the capture range of unconstrained convolutional filters. In the proposed pipeline, the Optimal Cross-section Matching (OCM) algorithm first performs constrained global alignment-translation, rotation, and uniform scaling-to establish anatomically consistent slice initialization. Next, a lightweight deep-learning refinement network, inspired by VoxelMorph, predicts residual local deformations between consecutive slices. The core novelty of this architecture lies in its hierarchical decomposition of the registration manifold: the OCM acts as a deterministic geometric anchor that neutralizes high-amplitude variance, thereby constraining the learning task to a low-dimensional residual manifold. This hybrid OCM + DL design integrates explicit geometric priors with the flexible learning capacity of neural networks, ensuring stable optimization and plausible deformation fields even with few training examples. Experiments on an original dataset of 40 kidneys demonstrated that the OCM + DL method achieved the highest registration accuracy across all evaluated metrics: NCC = 0.91, SSIM = 0.81, Dice = 0.90, IoU = 0.81, HD95 = 1.9 mm, and volumetric agreement DCVol = 0.89. Compared to single-stage baselines, this represents an average improvement of approximately 17% over DL-only and 14% over OCM-only, validating the synergistic contribution of the proposed hybrid strategy over standalone iterative or data-driven methods. The pipeline maintains physical calibration via Hough-based grid detection and employs B & eacute;zier-based contour smoothing for robust meshing and volume estimation. Although validated on kidney data, the proposed framework generalizes to other soft-tissue organs reconstructed from optical or photographic cross-sections. By decoupling interpretable global optimization from data-efficient deep refinement, the method advances the precision, reproducibility, and anatomical realism of multimodal 3D reconstructions for surgical planning, morphological assessment, and medical education.

Magnetic Resonance Imaging is increasing in importance in prostate cancer diagnosis due to the high accuracy and quality of the examination procedure. However, this process requires a time-consuming analysis of the results. Currently, machine vision is widely used in many areas. It enables automation and support in radiological studies. Successful detection of primary prostate tumors depends on the effective segmentation of the prostate itself. At times, a CT scan may be performed; alternatively, MRI may be the selected option. The data always reach a bottleneck stage. This paper presents the effective training of deep learning models to segment the prostate based on onefold and multimodal medical images. This approach supports the computer-aided diagnosis (CAD) system for radiologists as the first step in cancer exams. A comparison of two approaches designed for prostate segmentation is described. The first combines YOLOv4, the object detection neural network, and U-Net for a semantic segmentation based on onefold modality MRI images. The second presents the same method trained on multimodal images—a CT and MRI mixed dataset. The learning process was carried out in a cloud environment using GPU cards. The experiments are based on data from 120 patients who have undergone MRI and CT examinations. Several metrics evaluated the trained models. In the prostate semantic segmentation process, better results were achieved by mixed MRI with CT datasets. The best model achieved the value of 0.9685 for the Sørensen–Dice coefficient for the threshold value of 0.6.

This article presents a novel multiple organ localization and tracking technique applied to spleen and kidney regions in computed tomography images. The proposed solution is based on a unique approach to classify regions in different spatial projections (e.g., side projection) using convolutional neural networks. Our procedure merges classification results from different projection resulting in a 3D segmentation. The proposed system is able to recognize the contour of the organ with an accuracy of 88-89% depending on the body organ. Research has shown that the use of a single method can be useful for the detection of different organs: kidney and spleen. Our solution can compete with U-Net based solutions in terms of hardware requirements, as it has significantly lower demands. Additionally, it gives better results in small data sets. Another advantage of our solution is a significantly lower training time on an equally sized data set and more capabilities to parallelize calculations. The proposed system enables visualization, localization and tracking of organs and is therefore a valuable tool in medical diagnostic problems.

Image resizing is frequently used as a preprocessing step in many computer vision tasks, especially in medical applications. While tuning of the resizing method is usually omitted in the studies, there are many problems in which the exact influence of resampling on image textures and gradients is significant. The paper presents an in-depth analysis of image reconstruction's impact on two inherent tasks in medical image analysis: segmentation and classification. The proposed study is conducted on the renal diagnosis dataset in which the kidney is segmented, and three renal tumours are classified. A novel image reconstruction method is introduced, namely Sampling Kantorovich Algorithm (SKA). It is compared to six other popular techniques widely used in image processing. Based on the qualitative and quantitative analyses, we proved that choice of image reconstruction method impacts the system's overall performance. SKA turns out to be the best performing method in the classification setup. It boosts performance to 75% of the weighted F1-score by approximately 3 percentage points (pp) compared to the best baseline solution. In kidney segmentation, the SKA improves efficiency by over 2pp compared to other resizing methods. The results presented in this paper may apply to a wide range of medical image processing problems.

This paper presents the evaluation of the accurateness of an automatic HE to PAS stain conversion. We collected the unique HE-PAS (stain and restain specimens) database of renal specimens, and we have developed and compared a set of GAN methods for the automatic staining conversion. The detailed evaluation includes 10 numerical metrics and a visual evaluation, which was performed to investigate and closely look at the accuracy of an automatic stain transformation. The main contribution and novelty of this paper is the multi-aspect evaluation of the accurateness of staining transformation using a unique HE-PAS stain-restain dataset. Achieved results show that an automatic HE to PAS staining conversion achieved Frechet Inception Distance (FID) metric below 7, whereas visual evaluation is up to 96%. Presented analysis and results give insights into possibilities to build automatic stain transformations tools to replace the need of performing manual staining.

This work presents an automatic system for generating kidney boundaries in computed tomography (CT) images. This paper presents the main points of medical image processing, which are the parts of the developed system. The U-Net network was used for image segmentation, which is now widely used as a standard solution for many medical image processing tasks. An innovative solution for framing the input data has been implemented to improve the quality of the learning data as well as to reduce the size of the data. Precision-recall analysis was performed to calculate the optimal image threshold value. To eliminate false-positive errors, which are a common issue in segmentation based on neural networks, the volumetric analysis of coherent areas was applied. The developed system facilitates a fully automatic generation of kidney boundaries as well as the generation of a three-dimensional kidney model. The system can be helpful for people who deal with the analysis of medical images, medical specialists in medical centers, especially for those who perform the descriptions of CT examination. The system works fully automatically and can help to increase the accuracy of the performed medical diagnosis and reduce the time of preparing medical descriptions.

Objective: This article presents a novel method of automatic kidney contour detection in computed tomography angiography images. This technique allows to read as input the entire set of CTA images. It allows to read an entire set of kidney CTA images as input and then automatically generates binary images of the detected kidney outlines for each scan separately. Its additional feature is a real-time 3D kidney model reconstruction. Methods: The main idea is based on an innovative two-way scanning technique. To adapt an algorithm, a CT is analyzed on the basis of a previous slice. The final kidney contour recognition uses the following digital image processing techniques: mathematical morphology, region growth, colorization. Results: to assess the quality of our technique, we consulted the results with a pathology department. The F1 score of the researched method is 88 % compared to human specialist's verification. We also conducted a comparative study of computation time, system reliability, and recognition accuracy using three recent alternative methods. Conclusion: In comparison to machine learning algorithms, the presented method is very precise thanks to the application of the adaptive sweeping technique. This solution can be successfully applied in CTA image analyzing, visualization, and neoplastic changes detection. Significance: computer-aided medical diagnostic is currently one of the greatest challenges for biomedical engineers. The technique can find a real-life application in medical centers and medical-pathology departments.

The article presents an innovative technique of kidney detection in computed tomography images. The proposed system is based on a batch-based synthesis algorithm, never used in medical image processing before. The manuscript presents key points: U-Net-based initial segmentation, tracking system and image completion. The proposed system's main concept is to remove the kidney region from an image based on the previous image mask and calculate the difference between the original image and the image with the removed kidney. The developed system was compared to the patterns obtained by specialists from medical centers. Numerical tests were carried out confirming kidney detection's high effectiveness, which equals $90.63\pm 4.59\ \%$. The system's high performance and low hardware requirements compared to typical techniques based on deep learning systems are its main advantages. Numerical experiments supplied performance analysis and comparison with an alternative solution based on U-Net image semantic segmentation. A measurable effect of the developed techniques is the concept of the original and practical fully automatic kidney detection and segmentation system in computed tomography images.

The article presents an innovative method of scanning slices in computed tomography. The presented technique allows for automatically generation of three-dimensional projection, determined by X-Y-Z surfaces. Projections allow to calculate Kidney-Region-Of-Interest - a minimal envelope of kidney contours for all CT scans. The presented technique increases the accuracy of automatic identification and segmentation of kidneys and can be a good starting point for other techniques for identifying kidney areas on individual CT scans. This method significantly limits the area of kidney searching, thereby accelerates the operation of identification for any algorithm. The presented method is based on the technique of pixel intensity values averaging, area region-growing algorithms and morphological transformations. The presented technique has also been tested in implementation of the U-Net neural network system. Our presented solution of X-Y-Z projection is characterized by a high efficiency of visualization, comparable to the results obtained by a human expert.

The article presents an innovative method of 3D computer tomography (CT) image reconstruction of kidney. Diagnosis based on CT scanning allows to obtain projections of multi-dimensional object, made from different directions in order to create cross-sectional (2D) slices. Standard techniques for identifying kidneys in CT images analyze each 2D slice separately. It causes different reconstruction accuracy for the same object at its different heights. This is the main problem of a machine-learning systems. Reconstruction error of end-slices of the kidney model is often greater than the error of the kidney's middle part. The main idea of the technique presented in this paper is to analyze the largest coherent 3D spatial-areas. This technique allows to increase the accuracy of kidney detection as well as to decrease the FP (false positive) error. An additional advantage of the developed algorithm is the possibility of obtaining a precise model representing the 3D view of an entire kidney.

This article presents an innovative processing method of computed tomography image-slices for automatic identification of kidneys and cancers boundaries. The system was developed based on the use of U-Net convolutional neural network. The innovative solution proposed in the presented experiments uses the technique of automatically partitioning the input data into frames of fixed dimensions, containing only areas of interest. The developed technique allows obtaining accurate boundaries of the kidneys and cancers with high accuracy, comparable to a human expert. The obtained segmentation accuracy is satisfactory and equals 96 % for renal, 74% for renal cell cancers and 90 % for cystics.

The paper presents the deep learning ensemble of classifiers in recognition of melanoma on the basis of dermoscopy image analysis. The ensemble is based on 9 units supplied by the activation signals from the convolutional neural network. To provide the independence of unit operation few different feature selection methods combined with three types of classification networks have been used. The pre-trained Alexnet CNN structure has been used in this application. The experiments have been performed using two data bases in recognition of melanoma and non-melanoma cases. One of them is very well known large ISIC base and the second smaller data base collected in Warsaw Memorial Cancer Center and Institute of Oncology. The results have shown advantage of the ensemble over individually running classifiers. The accuracy was increased by few percentage points.

This article describes the automated computed tomography (CT) image processing technique supporting kidney detection. The main goal of the study is a fully automatic generation of a kidney boundary for each slice in the set of slices obtained in the computed tomography examination. This work describes three main tasks in the process of automatic kidney identification: the initial location of the kidneys using the U-Net convolutional neural network, the generation of an accurate kidney boundary using extended maxima transformation, and the application of the slice scanning algorithm supporting the process of generating the result for the next slice, using the result of the previous one. To assess the quality of the proposed technique of medical image analysis, automatic numerical tests were performed. In the test section, we presented numerical results, calculating the F1-score of kidney boundary detection by an automatic system, compared to the kidneys boundaries manually generated by a human expert from a medical center. The influence of the use of U-Net support in the initial detection of the kidney on the final F1-score of generating the kidney outline was also evaluated. The F1-score achieved by the automated system is 84% ± 10% for the system without U-Net support and 89% ± 9% for the system with U-Net support. Performance tests show that the presented technique can generate the kidney boundary up to 3 times faster than raw U-Net-based network. The proposed kidney recognition system can be successfully used in systems that require a very fast image processing time. The measurable effect of the developed techniques is a practical help for doctors, specialists from medical centers dealing with the analysis and description of medical image data.

This article presents the concept of a complex system for automatic detection of kidneys and kidney tumors in computed tomography images.An effective treatment of cancer depends on a quick and effective diagnosis.Computer support for medical diagnostics is crucial in effective specialists' analysis.Automatic and accurate location, together with precise detection of the kidney and/or tumor contour is a demanding task.In this article, authors present a complex system for automatic detection of kidneys and kidney tumors, based on machine learning techniques, using the U-Net network.Convolutional neural network recognition results are then processed in multiple stages, using morphological processing, 3D model analysis, geometric coefficients analysis and region-growth implementation.The results of the system detection were compared to the reference images marked by an expert.The system presented in the article is characterized by a very high efficiency of recognition and segmentation of kidney and tumor areas.

This paper presents a new image processing and analysis technique for the quality evaluation of cell nuclei to support medical diagnostics in breast cancer. The technique allows cell nuclei that are deformed or overlapped by biological material to be reconstructed. The paper proposes a sensitivity and similarity approach, enriching the PatchMatch correspondence algorithm in accurate cell reconstruction. Its application in reconstruction processes enables accelerated computations and an increased probability of obtaining appropriate segmentation results. The numerical results demonstrate that the developed system allows for automatic and effective cell nuclei reconstruction with an acceptable average area accuracy level above 85% compared with manual human results (assuming manual segmentation as a true value). The reconstruction system allows for the recovery of the proper shape of the analyzed distorted cells very rapidly and in a repeatable manner. An additional advantage of the procedure is that the nuclei area overlapped by artifacts or other cells can be determined. The experimental results prove the high utility of the method in final HER2 gene amplification assessment in breast cancer images. (C) 2019 Published by Elsevier Ltd.

The article presents an innovative method of kidney recognition in computed tomography (CT) images. Kidney cancer is one of the most common causes of death. Over 300,000 people die per year from this disease. A fast and correct diagnosis of neoplastic lesions in computed tomography images allows to choose the proper method of treatment. This article presents innovative and unique methods of kidney recognition in CT images. The proposed methods are based on morphological operations, shape analysis, geometrical coefficients calculations as well as the directional operation of flood fill with automatic selection of the stop criterion. The article presents also an innovative method of closing the boundary of an unrecognized kidney. Application of fast and effective algorithms for an automatic kidney shape recognition allows to make a 3D reconstruction of the kidney model. The use of algorithms to improve visualization of CT scans allows more accurate diagnosis by specialists. The system for supporting kidney cancer diagnosis presented in the article has been tested to assess the quality of kidney shape recognition. The recognition results of the shape of the kidney by the automatic system are comparable to the results obtained by a human expert and the accuracy of the diagnosis is at the level of 86%. Despite the difficult task, it was possible to obtain satisfactory results of the kidney shape recognition.

The article presents an original method of kidney recognition in CT images. Fast and precise visualization of a kidney is a very important task in medical diagnosis of kidney cancer. Calculation of a precise kidney's boundary allows to visualize any changes in the genitourinary system requiring specialist intervention. The method presented in this article uses shape analysis of objects' boundary and calculates geometrical coefficients for final detection. The results of the experiments show that the level of accuracy of automatic kidney boundary detection is 84 % comparing to expert's results.

The article presents a complex method of recognition nuclei cells areas and of segmentation of nuclei. The evaluation process of the identification and segmentation quality of proposed methods using L2 distance function and sensitivity function is also presented. FISH test is a fluorescence technique used for staining of microscope images of breast cancer. The technique allows visualization of HER2, CEN17 genes and cells nuclei. Fast and efficient microscopy image analysis allows a proper choice of therapy. This article presents a new, complex technique based on the color analysis, morphological transformations and watershed segmentation. The technique allows rapid and efficient identification of nuclei areas, as well as precise detection of the cells nuclei outlines. This step is often overlooked in a computer image analysis, whereas it is extremely important. It allows to increase the accuracy of HER2/CEN17 gene detection, as well as it allows to exclude fake biomarkers and increase the speed of identification of algorithms for HER2 genes by limiting the searched area. Proper segmentation of nuclei also makes manual evaluation of images easier.

The article presents an innovative approach to the automatic detection of cells nuclei in FISH microscopic images. The proposed solution is based on L2 distance function which is used in patterns recognition in images. The results of researches show that an efficient identification of cells nuclei in FISH images with the presented method is possible. The numerical results show that the accuracy of recognition performed by the automated system is at the level of 89% comparing to the expert's results.