Valery Starovoitov
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Valery Starovoitov has a total of 28 co-authors. Here are the top co-authors by number of shared publications:
- Marina Lukashevich
Information Management Systems Department Belarusian State University Minsk Belarus
2 shared publications U. Yu. Akhundjanov
Tashkent University of Information Technologies, Ferghana Branch, Ferghana, Uzbekistan
2 shared publications- Yu. I. Golub
United Institute of Informatics Problems, National Academy of Sciences of Belarus, Minsk, Belarus
2 shared publications - Yu. I. Golub
United Institute of Informatics Problems, National Academy of Sciences of Belarus, Minsk, Belarus
2 shared publications - Agnieszka Kitlas Golińska
University of Białystok
1 shared publication - Ana Gavrovska
School of Electrical Engineering, University of Belgrade, Bulevar kralja Aleksandra 73, 11120 Belgrade, Serbia
1 shared publication
Latest Publications
- Open Access
- 7 August 2026
- Published by MDPI AG in Information
- Vol. 17 (8) , 759
Building Detection Under Forest Canopy Using Physically Interpretable SAR Features and Hybrid Machine Learning Across Multiple Forest Biomes
Forests cover nearly one-third of the Earth’s land surface and are subject to increasing anthropogenic pressure, including unauthorised construction, infrastructure expansion, and habitat fragmentation. Detecting buildings concealed beneath forest canopy is essential for environmental monitoring and territorial surveillance, yet optical satellite imagery fails under persistent cloud cover and dense vegetation, and no existing approach provides reliable detection across contrasting forest ecosystems. We address this gap by proposing a hybrid SAR-based detection framework that combines twelve physically interpretable Scattering-Informed SAR Features (SISF)—derived from electromagnetic scattering theory across amplitude, polarimetric, temporal, and texture dimensions—with a universal Convolutional Neural Network, integrated through probability-level fusion with isotonic regional calibration. Rather than relying on individual feature thresholds, the framework identifies buildings through their characteristic multidimensional scattering signature—a combination that remains discriminative across biomes where any single SAR descriptor would fail. The framework was evaluated on a novel 7721-object multi-biome benchmark spanning forest-steppe (Kazakhstan), boreal forest (Komi Republic, Russia), and tropical rainforest (Brazil, Pará). The final Fusion + Regional Calibration model achieved an overall F1-score of 0.803 on the held-out test set (N = 1545), outperforming single-model baselines by up to 17 percentage points. Regional F1-scores ranged from 0.727 (Kazakhstan) to 0.944 (Komi Republic), with detection performance remaining robust under partial canopy occlusion (F1 = 0.911, versus 0.903 for unobscured buildings). An empirical inverse relationship between hard negative proportion and detection F1-score was identified within each biome, with the 28–35% proportions present in our sampled regions reported as a preliminary observation rather than a generally optimal range—a dataset design finding not previously reported in the literature. The proposed framework provides a physically interpretable solution for SAR-based building detection under forest canopy, demonstrating consistent performance across three contrasting forest biomes, with direct applications to environmental monitoring and territorial surveillance in forested regions.
- 7 August 2026
- Published by MDPI AG in Information
- Vol. 17 (8) , 759
Open AccessBuilding Detection Under Forest Canopy Using Physically Interpretable SAR Features and Hybrid Machine Learning Across Multiple Forest Biomes
Forests cover nearly one-third of the Earth’s land surface and are subject to increasing anthropogenic pressure, including unauthorised construction, infrastructure expansion, and habitat fragmentation. Detecting buildings concealed beneath forest canopy is essential for environmental monitoring and territorial surveillance, yet optical satellite imagery fails under persistent cloud cover and dense vegetation, and no existing approach provides reliable detection across contrasting forest ecosystems. We address this gap by proposing a hybrid SAR-based detection framework that combines twelve physically interpretable Scattering-Informed SAR Features (SISF)—derived from electromagnetic scattering theory across amplitude, polarimetric, temporal, and texture dimensions—with a universal Convolutional Neural Network, integrated through probability-level fusion with isotonic regional calibration. Rather than relying on individual feature thresholds, the framework identifies buildings through their characteristic multidimensional scattering signature—a combination that remains discriminative across biomes where any single SAR descriptor would fail. The framework was evaluated on a novel 7721-object multi-biome benchmark spanning forest-steppe (Kazakhstan), boreal forest (Komi Republic, Russia), and tropical rainforest (Brazil, Pará). The final Fusion + Regional Calibration model achieved an overall F1-score of 0.803 on the held-out test set (N = 1545), outperforming single-model baselines by up to 17 percentage points. Regional F1-scores ranged from 0.727 (Kazakhstan) to 0.944 (Komi Republic), with detection performance remaining robust under partial canopy occlusion (F1 = 0.911, versus 0.903 for unobscured buildings). An empirical inverse relationship between hard negative proportion and detection F1-score was identified within each biome, with the 28–35% proportions present in our sampled regions reported as a preliminary observation rather than a generally optimal range—a dataset design finding not previously reported in the literature. The proposed framework provides a physically interpretable solution for SAR-based building detection under forest canopy, demonstrating consistent performance across three contrasting forest biomes, with direct applications to environmental monitoring and territorial surveillance in forested regions.
- Open Access
- 30 December 2024
- Published by United Institute of Informatics Problems of the National Academy of Sciences of Belarus in Informatics
- Vol. 21 (4) , 72-84
Verification of normalized online signatures without calculating dynamic features
Objectives. Study of the method of verification of the authenticity of a human signature made on a tablet with a stylus and given three parameters: coordinates X, Y and pressure on the tablet P.Methods. N genuine dynamic human signatures are given. Data describing different signatures made by one person always have a different number of points. The main variants of normalization of the original signature data are investigated. A model of an individual image of human signatures is built without calculating dynamic features. The method of dynamic time transformation (DTW) is used to compare similar data of different signatures. The results of this transformation are DTW-distances between the data of pairs of signatures. These distances serve as coordinates of a point in the feature space describing the similarity of a pair of signatures. A set of such points represents a model describing the similarity of genuine human signatures. The parameters of the signature being verified are compared with each of the N authentic signatures used to build the model for their proximity to the model. If more than half of these pairs are further from the model than a certain threshold T, the signature is considered fake.Results. As a result of comparative experiments, a variant of normalization of initial data of dynamic human signatures was found, which allows verification of such signatures without calculating additional features usually calculated from the initial data X, Y, P.Conclusion. Experiments to generate individual signature models for each of 230 people from the publicly available MCYT dynamic signature database (a subset of the DeepSignDB database) and verify the authenticity of 11,500 signatures made on behalf of these people showed a verification accuracy of 99.82 %. Half of them were genuine, half were fake.
- 30 December 2024
- Published by United Institute of Informatics Problems of the National Academy of Sciences of Belarus in Informatics
- Vol. 21 (4) , 72-84
Open AccessVerification of normalized online signatures without calculating dynamic features
Objectives. Study of the method of verification of the authenticity of a human signature made on a tablet with a stylus and given three parameters: coordinates X, Y and pressure on the tablet P.Methods. N genuine dynamic human signatures are given. Data describing different signatures made by one person always have a different number of points. The main variants of normalization of the original signature data are investigated. A model of an individual image of human signatures is built without calculating dynamic features. The method of dynamic time transformation (DTW) is used to compare similar data of different signatures. The results of this transformation are DTW-distances between the data of pairs of signatures. These distances serve as coordinates of a point in the feature space describing the similarity of a pair of signatures. A set of such points represents a model describing the similarity of genuine human signatures. The parameters of the signature being verified are compared with each of the N authentic signatures used to build the model for their proximity to the model. If more than half of these pairs are further from the model than a certain threshold T, the signature is considered fake.Results. As a result of comparative experiments, a variant of normalization of initial data of dynamic human signatures was found, which allows verification of such signatures without calculating additional features usually calculated from the initial data X, Y, P.Conclusion. Experiments to generate individual signature models for each of 230 people from the publicly available MCYT dynamic signature database (a subset of the DeepSignDB database) and verify the authenticity of 11,500 signatures made on behalf of these people showed a verification accuracy of 99.82 %. Half of them were genuine, half were fake.
- 1 June 2024
- Published by Pleiades Publishing Ltd in Pattern Recognition and Image Analysis
- Vol. 34 (2) , 340-351
A Writer-Dependent Approach to Offline Signature Verification Based on One-Class Support Vector Machine
A new solution to the problem of offline signature verification is presented. Digital images of signatures are processed and converted into a binary representation of a certain size. Then their contours are traced, and from them, two original features are calculated for describing the local structural features of the signature in the form of vectors of normalized frequency distributions of local binary pattern codes and values of local curvature of the signature contours. A new feature space is formed in which the pattern describes the proximity of pairs of signatures, and its coordinates are the rank correlation coefficients between the feature vectors of these signatures. In real practice, the expert has M (from 5 to 15) genuine signatures of a person; there are no forged signatures at all. On these M available genuine signatures of a single person, we train a one-class support vector machine model and obtain a single-writer-dependent classifier. A verifiable signature is considered forged if the classifier model considers it to be an outlier. The accuracy of our approach in verifying the genuineness of all 2640 signatures from the CEDAR database was 99.77%. All forged signatures in this database were correctly recognized.
- 1 June 2024
- Published by Pleiades Publishing Ltd in Pattern Recognition and Image Analysis
- Vol. 34 (2) , 340-351
A Writer-Dependent Approach to Offline Signature Verification Based on One-Class Support Vector Machine
A new solution to the problem of offline signature verification is presented. Digital images of signatures are processed and converted into a binary representation of a certain size. Then their contours are traced, and from them, two original features are calculated for describing the local structural features of the signature in the form of vectors of normalized frequency distributions of local binary pattern codes and values of local curvature of the signature contours. A new feature space is formed in which the pattern describes the proximity of pairs of signatures, and its coordinates are the rank correlation coefficients between the feature vectors of these signatures. In real practice, the expert has M (from 5 to 15) genuine signatures of a person; there are no forged signatures at all. On these M available genuine signatures of a single person, we train a one-class support vector machine model and obtain a single-writer-dependent classifier. A verifiable signature is considered forged if the classifier model considers it to be an outlier. The accuracy of our approach in verifying the genuineness of all 2640 signatures from the CEDAR database was 99.77%. All forged signatures in this database were correctly recognized.
- Open Access
- 16 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 77092-77103
A Universal Field-of-View Mask Segmentation Method on Retinal Images From Fundus Cameras
One of the first steps in the retinal image preprocessing is cropping the Field of View (FOV) area and scaling it into a template of a predefined size. Fundus cameras of different producers record digital images of the retina of various sizes, and the FOV area containing helpful information can be from 43 to 98% of the image area. For automated analysis of retinal images and detection of DR, it is necessary to segment the FOV region and cut it out from the image. This is important to preserve microaneurysms and small capillaries in the retinal image as much as possible, since neural network methods always reduce the original images to a predefined size. In this paper, we propose a universal method for FOV segmentation based on the ideas of histograms and thresholds. We compared 11 methods for segmenting FOV regions on the four most commonly used retinal image grayscale representations. In total, we compared 35 variants of segmentation and evaluated the obtained results by four functions: Jaccard index, Matthews correlation coefficient (MCC), accuracy and balanced accuracy. All options were tested on 7000 images from eight of the largest databases. The images were generated by 100 different fundus cameras. The following observations have been extracted through extensive comparative experiments namely: 1) segmentation of the FOV area should be performed on the grayscale image obtained from the red channel; 2) for more accurate segmentation, a logarithmic transformation should be applied to the grayscale image; 3) the FOV area mask can be segmented by a global threshold calculated by Otsu’s method; 4) global thresholding based on analysis of histogram peaks does not provide advantages over binarization by Otsu’s method applied to the logarithmic transformation of the image.
- 16 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 77092-77103
Open AccessA Universal Field-of-View Mask Segmentation Method on Retinal Images From Fundus Cameras
One of the first steps in the retinal image preprocessing is cropping the Field of View (FOV) area and scaling it into a template of a predefined size. Fundus cameras of different producers record digital images of the retina of various sizes, and the FOV area containing helpful information can be from 43 to 98% of the image area. For automated analysis of retinal images and detection of DR, it is necessary to segment the FOV region and cut it out from the image. This is important to preserve microaneurysms and small capillaries in the retinal image as much as possible, since neural network methods always reduce the original images to a predefined size. In this paper, we propose a universal method for FOV segmentation based on the ideas of histograms and thresholds. We compared 11 methods for segmenting FOV regions on the four most commonly used retinal image grayscale representations. In total, we compared 35 variants of segmentation and evaluated the obtained results by four functions: Jaccard index, Matthews correlation coefficient (MCC), accuracy and balanced accuracy. All options were tested on 7000 images from eight of the largest databases. The images were generated by 100 different fundus cameras. The following observations have been extracted through extensive comparative experiments namely: 1) segmentation of the FOV area should be performed on the grayscale image obtained from the red channel; 2) for more accurate segmentation, a logarithmic transformation should be applied to the grayscale image; 3) the FOV area mask can be segmented by a global threshold calculated by Otsu’s method; 4) global thresholding based on analysis of histogram peaks does not provide advantages over binarization by Otsu’s method applied to the logarithmic transformation of the image.
- Open Access
- 1 January 2023
- Published by Centre for Evaluation in Education and Science (CEON/CEES) in Telfor Journal
- Vol. 15 (2) , 32-37
Quadratic fitting model in no-reference image quality assessment
The perceptual quality of image is affected by distortions during compression, delivery and storage. Distortions also impact automatic image quality assessment (IQA) that needs to be highly correlated with subjective scores. In the absence of reference, which is a typical scenario in practice, no-reference (NR) metrics are necessary for quality measurements. Recently such methods are proposed, and they employ natural scene statistics (NSS). The experimental analysis performed in this paper takes into consideration two fitting or regression models of several NR-IQA metrics relying on different distortion types. The results show quadratic model as promising for making relations in terms of difference mean opinion score and Shannon entropy.
- 1 January 2023
- Published by Centre for Evaluation in Education and Science (CEON/CEES) in Telfor Journal
- Vol. 15 (2) , 32-37
Open AccessQuadratic fitting model in no-reference image quality assessment
The perceptual quality of image is affected by distortions during compression, delivery and storage. Distortions also impact automatic image quality assessment (IQA) that needs to be highly correlated with subjective scores. In the absence of reference, which is a typical scenario in practice, no-reference (NR) metrics are necessary for quality measurements. Recently such methods are proposed, and they employ natural scene statistics (NSS). The experimental analysis performed in this paper takes into consideration two fitting or regression models of several NR-IQA metrics relying on different distortion types. The results show quadratic model as promising for making relations in terms of difference mean opinion score and Shannon entropy.
- Open Access
- 22 August 2022
- Published by United Institute of Informatics Problems of the National Academy of Sciences of Belarus in Informatics
- Vol. 19 (3) , 62-73
A new feature for handwritten signature image description based on local binary patterns
Objectives. The problem of describing the invariant features of a digital image of handwritten signature that describes the distribution of its local features is considered. The formation of fundamentally new approach to the calculation of such features is described.Methods. Digital image processing methods are used. First an image is converted into a binary representation, then its morphological and median filtering is performed. Then using the method of principal components, the image is rotated to give the signature a horizontal orientation. A rectangle describing the signature is cut out, then it is scaled to the template of a certain size. In the article the template of 300×150 pixels was used. Then the border of the signature is formed. Local binary patterns are calculated from its binary contour, i.e. each pixel is assigned a number from 0 to 255, which describes the location of the edge pixels in 3×3 neighborhood of each pixel. A histogram of calculated patterns for 256 intervals is formed. The first and last intervals are discarded because they correspond to all black and white pixels in the neighborhood and are not informative. The remaining 254 numbers of the array form new local features of the signature.Results. The studies were performed on the bases of digitized signatures TUIT and CEDAR containing true and fake signatures of 80 persons. The accuracy of correct verification of signatures on these bases was about 78 % and 70 %.Conclusion. The possibility of using the proposed possibilities for solving the problems of verifying the authenticity of handwritten signatures has been experimentally confirmed.
- 22 August 2022
- Published by United Institute of Informatics Problems of the National Academy of Sciences of Belarus in Informatics
- Vol. 19 (3) , 62-73
Open AccessA new feature for handwritten signature image description based on local binary patterns
Objectives. The problem of describing the invariant features of a digital image of handwritten signature that describes the distribution of its local features is considered. The formation of fundamentally new approach to the calculation of such features is described.Methods. Digital image processing methods are used. First an image is converted into a binary representation, then its morphological and median filtering is performed. Then using the method of principal components, the image is rotated to give the signature a horizontal orientation. A rectangle describing the signature is cut out, then it is scaled to the template of a certain size. In the article the template of 300×150 pixels was used. Then the border of the signature is formed. Local binary patterns are calculated from its binary contour, i.e. each pixel is assigned a number from 0 to 255, which describes the location of the edge pixels in 3×3 neighborhood of each pixel. A histogram of calculated patterns for 256 intervals is formed. The first and last intervals are discarded because they correspond to all black and white pixels in the neighborhood and are not informative. The remaining 254 numbers of the array form new local features of the signature.Results. The studies were performed on the bases of digitized signatures TUIT and CEDAR containing true and fake signatures of 80 persons. The accuracy of correct verification of signatures on these bases was about 78 % and 70 %.Conclusion. The possibility of using the proposed possibilities for solving the problems of verifying the authenticity of handwritten signatures has been experimentally confirmed.
- 1 June 2022
- Published by Pleiades Publishing Ltd in Pattern Recognition and Image Analysis
- Vol. 32 (2) , 322-331
A Universal Retinal Image Template for Automated Screening of Diabetic Retinopathy
Diabetic retinopathy (DR) frequently appears in diabetic patients. It is initially asymptomatic, but can progress to blindness. Screening studies for its diagnosis are performed in many countries by means of photographing the eye retina with special fundus-cameras. These studies are aimed at revealing the presence of microaneurysms (MAs) on the retina, which are the primary signs of DR. The wide variety of cameras, peculiarities of retina illumination, FOV angles, and sizes of digital images has complicated the development of a reliable and universal approach to analyzing retina images by machine-learning methods. In this paper, we consider the problem of choosing the size and shape of a unified template for representing the data of an arbitrary retinal image for subsequent automated DR screening. It is experimentally proved that it is possible to extract a square inscribed in the FOV region from each retinal image and compress it to the size of 512 × 512 pixels. This is the minimum allowable size of the template. It preserves the required number of MAs for DR screening by machine-learning methods.
- 1 June 2022
- Published by Pleiades Publishing Ltd in Pattern Recognition and Image Analysis
- Vol. 32 (2) , 322-331
A Universal Retinal Image Template for Automated Screening of Diabetic Retinopathy
Diabetic retinopathy (DR) frequently appears in diabetic patients. It is initially asymptomatic, but can progress to blindness. Screening studies for its diagnosis are performed in many countries by means of photographing the eye retina with special fundus-cameras. These studies are aimed at revealing the presence of microaneurysms (MAs) on the retina, which are the primary signs of DR. The wide variety of cameras, peculiarities of retina illumination, FOV angles, and sizes of digital images has complicated the development of a reliable and universal approach to analyzing retina images by machine-learning methods. In this paper, we consider the problem of choosing the size and shape of a unified template for representing the data of an arbitrary retinal image for subsequent automated DR screening. It is experimentally proved that it is possible to extract a square inscribed in the FOV region from each retinal image and compress it to the size of 512 × 512 pixels. This is the minimum allowable size of the template. It preserves the required number of MAs for DR screening by machine-learning methods.
- 18 March 2022
- Published by Springer Nature
- p. 152-165
Retinal Image Analysis Approach for Diabetic Retinopathy Grading
- 18 March 2022
- Published by Springer Nature
- p. 152-165
Retinal Image Analysis Approach for Diabetic Retinopathy Grading
- Open Access
- 21 July 2021
- Published by Wiley in International Journal of Optics
- Vol. 2021, 1-12
Optic Disc and Optic Cup Segmentation for Glaucoma Detection from Blur Retinal Images Using Improved Mask-RCNN
Glaucoma is a fatal eye disease that harms the optic disc (OD) and optic cup (OC) and results into blindness in progressed phases. Because of slow progress, the disease exhibits a small number of symptoms in the initial stages, therefore causing the disease identification to be a complicated task. So, a fully automatic framework is mandatory, which can support the screening process and increase the chances of disease detection in the early stages. In this paper, we deal with the localization and segmentation of the OD and OC for glaucoma detection from blur retinal images. We have presented a novel method that is Densenet-77-based Mask-RCNN to overcome the challenges of the glaucoma detection. Initially, we have performed the data augmentation step together with adding blurriness in samples to increase the diversity of data. Then, we have generated the annotations from ground-truth (GT) images. After that, the Densenet-77 framework is employed at the feature extraction level of Mask-RCNN to compute the deep key points. Finally, the calculated features are used to localize and segment the OD and OC by the custom Mask-RCNN model. For performance evaluation, we have used the ORIGA dataset that is publicly available. Furthermore, we have performed cross-dataset validation on the HRF database to show the robustness of the presented framework. The presented framework has achieved an average precision, recall, F-measure, and IOU as 0.965, 0.963, 0.97, and 0.972, respectively. The proposed method achieved remarkable performance in terms of both efficiency and effectiveness as compared to the latest techniques under the presence of blurring, noise, and light variations.
- 21 July 2021
- Published by Wiley in International Journal of Optics
- Vol. 2021, 1-12
Open AccessOptic Disc and Optic Cup Segmentation for Glaucoma Detection from Blur Retinal Images Using Improved Mask-RCNN
Glaucoma is a fatal eye disease that harms the optic disc (OD) and optic cup (OC) and results into blindness in progressed phases. Because of slow progress, the disease exhibits a small number of symptoms in the initial stages, therefore causing the disease identification to be a complicated task. So, a fully automatic framework is mandatory, which can support the screening process and increase the chances of disease detection in the early stages. In this paper, we deal with the localization and segmentation of the OD and OC for glaucoma detection from blur retinal images. We have presented a novel method that is Densenet-77-based Mask-RCNN to overcome the challenges of the glaucoma detection. Initially, we have performed the data augmentation step together with adding blurriness in samples to increase the diversity of data. Then, we have generated the annotations from ground-truth (GT) images. After that, the Densenet-77 framework is employed at the feature extraction level of Mask-RCNN to compute the deep key points. Finally, the calculated features are used to localize and segment the OD and OC by the custom Mask-RCNN model. For performance evaluation, we have used the ORIGA dataset that is publicly available. Furthermore, we have performed cross-dataset validation on the HRF database to show the robustness of the presented framework. The presented framework has achieved an average precision, recall, F-measure, and IOU as 0.965, 0.963, 0.97, and 0.972, respectively. The proposed method achieved remarkable performance in terms of both efficiency and effectiveness as compared to the latest techniques under the presence of blurring, noise, and light variations.
- Open Access
- 24 March 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 9, 47112-47124
The Benefits of the Matthews Correlation Coefficient (MCC) Over the Diagnostic Odds Ratio (DOR) in Binary Classification Assessment
To assess the quality of a binary classification, researchers often take advantage of a four-entry contingency table called confusion matrix , containing true positives, true negatives, false positives, and false negatives. To recap the four values of a confusion matrix in a unique score, researchers and statisticians have developed several rates and metrics. In the past, several scientific studies already showed why the Matthews correlation coefficient (MCC) is more informative and trustworthy than confusion-entropy error, accuracy, F 1 score, bookmaker informedness, markedness, and balanced accuracy. In this study, we compare the MCC with the diagnostic odds ratio (DOR), a statistical rate employed sometimes in biomedical sciences. After examining the properties of the MCC and of the DOR, we describe the relationships between them, by also taking advantage of an innovative geometrical plot called confusion tetrahedron , presented here for the first time. We then report some use cases where the MCC and the DOR produce discordant outcomes, and explain why the Matthews correlation coefficient is more informative and reliable between the two. Our results can have a strong impact in computer science and statistics, because they clearly explain why the trustworthiness of the information provided by the Matthews correlation coefficient is higher than the one generated by the diagnostic odds ratio.
- 24 March 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 9, 47112-47124
Open AccessThe Benefits of the Matthews Correlation Coefficient (MCC) Over the Diagnostic Odds Ratio (DOR) in Binary Classification Assessment
To assess the quality of a binary classification, researchers often take advantage of a four-entry contingency table called confusion matrix , containing true positives, true negatives, false positives, and false negatives. To recap the four values of a confusion matrix in a unique score, researchers and statisticians have developed several rates and metrics. In the past, several scientific studies already showed why the Matthews correlation coefficient (MCC) is more informative and trustworthy than confusion-entropy error, accuracy, F 1 score, bookmaker informedness, markedness, and balanced accuracy. In this study, we compare the MCC with the diagnostic odds ratio (DOR), a statistical rate employed sometimes in biomedical sciences. After examining the properties of the MCC and of the DOR, we describe the relationships between them, by also taking advantage of an innovative geometrical plot called confusion tetrahedron , presented here for the first time. We then report some use cases where the MCC and the DOR produce discordant outcomes, and explain why the Matthews correlation coefficient is more informative and reliable between the two. Our results can have a strong impact in computer science and statistics, because they clearly explain why the trustworthiness of the information provided by the Matthews correlation coefficient is higher than the one generated by the diagnostic odds ratio.