An explainable Artificial Intelligence software system for predicting diabetes.

Implementing diabetes surveillance systems is paramount to mitigate the risk of incurring substantial medical expenses. Currently, blood glucose is measured by minimally invasive methods, which involve extracting a small blood sample and transmitting it to a blood glucose meter. This method is deemed discomforting for individuals who are undergoing it. The present study introduces an Explainable Artificial Intelligence (XAI) system, which aims to create an intelligible machine capable of explaining expected outcomes and decision models. To this end, we analyze abnormal glucose levels by utilizing Bi-directional Long Short-Term Memory (Bi-LSTM) and Convolutional Neural Network (CNN). In this regard, the glucose levels are acquired through the glucose oxidase (GOD) strips placed over the human body. Later, the signal data is converted to the spectrogram images, classified as low glucose, average glucose, and abnormal glucose levels. The labeled spectrogram images are then used to train the individualized monitoring model. The proposed XAI model to track real-time glucose levels uses the XAI-driven architecture in its feature processing. The model's effectiveness is evaluated by analyzing the performance of the proposed model and several evolutionary metrics used in the confusion matrix. The data revealed in the study demonstrate that the proposed model effectively identifies individuals with elevated glucose levels.

Comparative Analysis of Water Stress Regimes in Avocado Plants during the Early Development Stage.

The avocado cv. Hass requires a suitable rootstock for optimal development under water stress. This study evaluated the performance of two avocado rootstocks (ANRR88 and ANGI52) grafted onto cv. Hass under four water stress conditions, 50% and 25% deficit, and 50% and 25% excess during the nursery stage. Plant height, leaf area (LA), dry matter (DM), and Carbon (OC) content in the roots, stems, and leaves were measured. Root traits were evaluated using digital imaging, and three vegetation indices (NDVI, CIRE, and MTCI) were used to quantify stress. The results showed that genotype significantly influenced the response to water stress. ANRR88 exhibited adaptation to moderate to high water deficits. ANGI52 adapted better to both water deficit and excess, and showed greater root exploration. LA and DM reductions of up to 60% were observed in ANRR88, suggesting a higher sensitivity to extreme changes in water availability. More than 90% of the total OC accumulation was observed in the stem and roots. The NDVI and the MTCI quantified the presence and levels of stress applied, and the 720 nm band provided high precision and speed for detecting stress. These insights are crucial for selecting rootstocks that ensure optimal performance under varying water availability, enhancing productivity and sustainability.

A Deep Learning Approach to Distance Map Generation Applied to Automatic Fiber Diameter Computation from Digital Micrographs.

Precise measurement of fiber diameter in animal and synthetic textiles is crucial for quality assessment and pricing; however, traditional methods often struggle with accuracy, particularly when fibers are densely packed or overlapping. Current computer vision techniques, while useful, have limitations in addressing these challenges. This paper introduces a novel deep-learning-based method to automatically generate distance maps of fiber micrographs, enabling more accurate fiber segmentation and diameter calculation. Our approach utilizes a modified U-Net architecture, trained on both real and simulated micrographs, to regress distance maps. This allows for the effective separation of individual fibers, even in complex scenarios. The model achieves a mean absolute error (MAE) of 0.1094 and a mean square error (MSE) of 0.0711, demonstrating its effectiveness in accurately measuring fiber diameters. This research highlights the potential of deep learning to revolutionize fiber analysis in the textile industry, offering a more precise and automated solution for quality control and pricing.

Significance of AI-assisted techniques for epiphyte plant monitoring and identification from drone images.

Monitoring forest canopies is vital for ecological studies, particularly for assessing epiphytes in rain forest ecosystems. Traditional methods for studying epiphytes, such as climbing trees and building observation structures, are labor, cost intensive and risky. Unmanned Aerial Vehicles (UAVs) have emerged as a valuable tool in this domain, offering botanists a safer and more cost-effective means to collect data. This study leverages AI-assisted techniques to enhance the identification and mapping of epiphytes using UAV imagery. The primary objective of this research is to evaluate the effectiveness of AI-assisted methods compared to traditional approaches in segmenting/identifying epiphytes from UAV images collected in a reserve forest in Costa Rica. Specifically, the study investigates whether Deep Learning (DL) models can accurately identify epiphytes during complex backgrounds, even with a limited dataset of varying image quality. Systematically, this study compares three traditional image segmentation methods Auto Cluster, Watershed, and Level Set with two DL-based segmentation networks: the UNet and the Vision Transformer-based TransUNet. Results obtained from this study indicate that traditional methods struggle with the complexity of vegetation backgrounds and variability in target characteristics. Epiphyte identification results were quantitatively evaluated using the Jaccard score. Among traditional methods, Watershed scored 0.10, Auto Cluster 0.13, and Level Set failed to identify the target. In contrast, AI-assisted models performed better, with UNet scoring 0.60 and TransUNet 0.65. These results highlight the potential of DL approaches to improve the accuracy and efficiency of epiphyte identification and mapping, advancing ecological research and conservation.

Semantic Segmentation in Large-Size Orthomosaics to Detect the Vegetation Area in Opuntia spp. Crop.

This study focuses on semantic segmentation in crop Opuntia spp. orthomosaics; this is a significant challenge due to the inherent variability in the captured images. Manual measurement of Opuntia spp. vegetation areas can be slow and inefficient, highlighting the need for more advanced and accurate methods. For this reason, we propose to use deep learning techniques to provide a more precise and efficient measurement of the vegetation area. Our research focuses on the unique difficulties posed by segmenting high-resolution images exceeding 2000 pixels, a common problem in generating orthomosaics for agricultural monitoring. The research was carried out on a Opuntia spp. cultivation located in the agricultural region of Tulancingo, Hidalgo, Mexico. The images used in this study were obtained by drones and processed using advanced semantic segmentation architectures, including DeepLabV3+, UNet, and UNet Style Xception. The results offer a comparative analysis of the performance of these architectures in the semantic segmentation of Opuntia spp., thus contributing to the development and improvement of crop analysis techniques based on deep learning. This work sets a precedent for future research applying deep learning techniques in agriculture.

Usefulness of Tissue Biomarkers versus Prostate-Specific Membrane Antigen-Positron Emission Tomography for Prostate Cancer Biochemical Recurrence after Radical Prostatectomy.

Despite curative-intent local therapy, approximately 27% to 53% of prostate cancer (PCa) patients experience prostate-specific antigen (PSA) recurrence, known as biochemical recurrence (BCR). BCR significantly raises the risk of PCa-related morbidity and mortality, yet there is no consensus on optimal management. Prostate-specific membrane antigen-positron emission tomography (PSMA PET) has emerged as highly sensitive imaging, distinguishing local recurrences from distant metastases, crucially influencing treatment decisions. Genomic biomarkers such as Decipher, Prolaris, and Oncotype DX contribute to refining recurrence risk profiles, guiding decisions on intensifying adjuvant therapies, like radiotherapy and androgen deprivation therapy (ADT). This review assesses PSMA PET and biomarker utility in post-radical prostatectomy BCR scenarios, highlighting their impact on clinical decision-making. Despite their promising roles, the routine integration of biomarkers is limited by availability and cost, requiring further evidence. PSMA PET remains indispensable for restaging and treatment evaluation in these patients. Integrating biomarkers and PSMA PET promises to optimize personalized management strategies for BCR, though more comprehensive consensus-building studies are needed to define their standardized utility in clinical practice.

Sagittal cranial projections dataset extracted from CBCT volumes corrected to natural head position in a mixed ethnicity population.

This dataset features 200 sagittal projection images derived from Cone Beam Computed Tomography (CBCT) scans, corrected according to the Natural Head Position (NHP) guidelines proposed by Fredrik Lundström and Anders Lundström. The images originate from orthodontic patients in Cali, Valle del Cauca, Colombia, encompassing both initial phases and ongoing treatments. The dataset is divided into two groups: 100 images from female subjects (CoF) and 100 from male subjects (CoM), facilitating gender-specific studies. The dataset is accompanied by an Excel file ``Data info.xlsx'' that details the rotation angles in the axial (Yaw), coronal (Roll), and sagittal (Pitch) planes, along with the pixel size and image dimensions. This detailed documentation supports the replication of studies and aids in the interpretation of cephalometric analyses. Corrections made to align the images with NHP standards involve adjustments in the three main anatomical planes using points from the frontozygomatic suture (Fz) in the axial and coronal planes, and sella (S) and nasion (N) for the sagittal plane.

Case report: Myelitis and ganglionitis, an atypical presentation of Hansen's disease.

Hansen's disease, or leprosy, is a disease characterized by dermatological and neurological disorders. A neural form also exists, in which peripheral neuropathy occurs in the absence of skin lesions. However, cases of leprosy that involve the central nervous system and proximal nerves are rare in the literature. We describe the case of an oligosymptomatic patient diagnosed with the neural form of leprosy with involvement of peripheral nerves, dorsal root ganglion, and cervical spinal cord in an atypical presentation of the disease. Through complementary examinations and nerve biopsies, the bacillus was identified, and treatment was subsequently initiated. This case highlights the importance of investigating the suspicion of leprosy, even in cases with atypical manifestations, as early diagnosis and treatment can reduce neurological damage and deformities.

[Cirrhotic cardiomyopathy ­ Clinically fact or academic curiosity? Review. Part 2: ECG, functional tests, images, biomarkers, screening for coronary heart disease and differentianting diagnosis] / Cardiomiopatía cirrótica ­ ¿Realidad clínica o simple curiosidad académica? Revisión. Parte 2: ECG, pruebas funcionales, estudios con imágenes, biomarcadores, screening para enfermedad coronaria y diagnósticos diferenciales.

The diagnosis of Cirrhotic Cardiomyopathy is based on severe hepatic cirrosis with deterioration of cardiac function without previous cardiopathy, but this is subclinical during a long time. In this second part we review the non-invasive diagnostic methods and their prognostic value in patients with or without hepatic transplant, from ECG to cardiac images of magnetic resonance.

El diagnóstico de Cardiomiopatía Cirrótica está basado en la presencia de cirrosis hepática avanzada con alteraciones de la función cardíaca sin cardiopatía pre-existente, pero en gran parte de su evolución natural ésta es subclínica. Por ello son imprescindibles los estudios complementarios no invasivos para confirmar el diagnóstico y su rol pronóstico en pacientes con o sin trasplante hepático. En esta segunda parte revisamos los métodos de diagnóstico desde el ECG hasta las imágenes de resonancia magnética cardíaca.

Redescription of the enigmatic myriapod Hanseniellachilensis (Hansen, 1903) (Symphyla, Scutigerellidae) based on scanning electron microscope images of Chilean specimens.

Hanseniellachilensis is the only myriapod of the class Symphyla known from Chile. This garden centipede, or pseudocentipede, was described more than 120 years ago based on morphologically incomplete specimens collected in central Chile, a well-known biodiversity hotspot. In this study, we redescribe this species based on morphologically complete specimens collected near the type locality using scanning electron microscope images. Our study provides the description of diagnostic characters hitherto unknown in this species such as macrochaetae of the tergites and spinnerets of the cerci. We also include a new record from central Chile and discuss the presumed presence of this species in Argentina and Madagascar.

Combining multisensor images and social network data to assess the area flooded by a hurricane event.

In this study, multisensor remote sensing datasets were used to characterize the land use and land covers (LULC) flooded by Hurricane Willa which made landfall on October 24, 2018. The landscape characterization was done using an unsupervised K-means algorithm of a cloud-free Sentinel-2 MultiSpectral Instrument (MSI) image, acquired during the dry season before Hurricane Willa. A flood map was derived using the histogram thresholding technique over a Synthetic Aperture Radar (SAR) Sentinel-1 C-band and combined with a flood map derived from a Sentinel-2 MSI image. Both, the Sentinel-1 and Sentinel-2 images were obtained after Willa landfall. While the LULC map reached an accuracy of 92%, validated using data collected during field surveys, the flood map achieved 90% overall accuracy, validated using locations extracted from social network data, that were manually georeferenced. The agriculture class was the dominant land use (about 2,624 km2), followed by deciduous forest (1,591 km2) and sub-perennial forest (1,317 km2). About 1,608 km2 represents the permanent wetlands (mangrove, salt marsh, lagoon and estuaries, and littoral classes), but only 489 km2 of this area belongs to aquatic surfaces (lagoons and estuaries). The flooded area was 1,225 km2, with the agricultural class as the most impacted (735 km2). Our analysis detected the saltmarsh class occupied 541 km2in the LULC map, and around 328 km2 were flooded during Hurricane Willa. Since the water flow receded relatively quickly, obtaining representative imagery to assess the flood event was a challenge. Still, the high overall accuracies obtained in this study allow us to assume that the outputs are reliable and can be used in the implementation of effective strategies for the protection, restoration, and management of wetlands. In addition, they will improve the capacity of local governments and residents of Marismas Nacionales to make informed decisions for the protection of vulnerable areas to the different threats derived from climate change.

Comprehensive analysis of clinical images contributions for melanoma classification using convolutional neural networks.

BACKGROUND: Timely diagnosis plays a critical role in determining melanoma prognosis, prompting the development of deep learning models to aid clinicians. Questions persist regarding the efficacy of clinical images alone or in conjunction with dermoscopy images for model training. This study aims to compare the classification performance for melanoma of three types of CNN models: those trained on clinical images, dermoscopy images, and a combination of paired clinical and dermoscopy images from the same lesion. MATERIALS AND METHODS: We divided 914 image pairs into training, validation, and test sets. Models were built using pre-trained Inception-ResNetV2 convolutional layers for feature extraction, followed by binary classification. Training comprised 20 models per CNN type using sets of random hyperparameters. Best models were chosen based on validation AUC-ROC. RESULTS: Significant AUC-ROC differences were found between clinical versus dermoscopy models (0.661 vs. 0.869, p < 0.001) and clinical versus clinical + dermoscopy models (0.661 vs. 0.822, p = 0.001). Significant sensitivity differences were found between clinical and dermoscopy models (0.513 vs. 0.799, p = 0.01), dermoscopy versus clinical + dermoscopy models (0.799 vs. 1.000, p = 0.02), and clinical versus clinical + dermoscopy models (0.513 vs. 1.000, p < 0.001). Significant specificity differences were found between dermoscopy versus clinical + dermoscopy models (0.800 vs. 0.288, p < 0.001) and clinical versus clinical + dermoscopy models (0.650 vs. 0.288, p < 0.001). CONCLUSION: CNN models trained on dermoscopy images outperformed those relying solely on clinical images under our study conditions. The potential advantages of incorporating paired clinical and dermoscopy images for CNN-based melanoma classification appear less clear based on our findings.

FPGA-Microprocessor Based Sensor for Faults Detection in Induction Motors Using Time-Frequency and Machine Learning Methods.

Induction motors (IM) play a fundamental role in the industrial sector because they are robust, efficient, and low-cost machines. Changes in the environment, installation errors, or modifications to working conditions can generate faults in induction motors. The trend on IM fault detection is focused on the design techniques and sensors capable of evaluating multiple faults with various signals using non-invasive analysis. The methodology is based on processing electric current signals by applying the short-time Fourier transform (STFT). Additionally, the computation of the mean and standard deviation of infrared thermograms is proposed as main indicators. The proposed system combines both parameters by means of Support Vector Machine and k-nearest-neighbor classifiers. The development of the diagnostic system was done with digital hardware implementations using a Xilinx PYNQ Z2 card that integrates an FPGA with a microprocessor, thus taking advantage of the acquisition and processing of digital signals and images in hardware. The proposed method has proved to be effective for the classification of healthy (HLT), misalignment (MAMT), unbalance (UNB), damaged bearing (BDF), and broken rotor bar (BRB) faults with an accuracy close to 99%.

Breast density quantification in dual-energy mammography using virtual anthropomorphic phantoms.

PURPOSE: Breast density is a significant risk factor for breast cancer and can impact the sensitivity of screening mammography. Area-based breast density measurements may not provide an accurate representation of the tissue distribution, therefore volumetric breast density (VBD) measurements are preferred. Dual-energy mammography enables volumetric measurements without additional assumptions about breast shape. In this work we evaluated the performance of a dual-energy decomposition technique for determining VBD by applying it to virtual anthropomorphic phantoms. METHODS: The dual-energy decomposition formalism was used to quantify VBD on simulated dual-energy images of anthropomorphic virtual phantoms with known tissue distributions. We simulated 150 phantoms with volumes ranging from 50 to 709 mL and VBD ranging from 15% to 60%. Using these results, we validated a correction for the presence of skin and assessed the method's intrinsic bias and variability. As a proof of concept, the method was applied to 14 sets of clinical dual-energy images, and the resulting breast densities were compared to magnetic resonance imaging (MRI) measurements. RESULTS: Virtual phantom VBD measurements exhibited a strong correlation (Pearson's r > 0.95 $r > 0.95$ ) with nominal values. The proposed skin correction eliminated the variability due to breast size and reduced the bias in VBD to a constant value of -2%. Disagreement between clinical VBD measurements using MRI and dual-energy mammography was under 10%, and the difference in the distributions was statistically non-significant. VBD measurements in both modalities had a moderate correlation (Spearman's ρ $\rho \ $ = 0.68). CONCLUSIONS: Our results in virtual phantoms indicate that the material decomposition method can produce accurate VBD measurements if the presence of a third material (skin) is considered. The results from our proof of concept showed agreement between MRI and dual-energy mammography VBD. Assessment of VBD using dual-energy images could provide complementary information in dual-energy mammography and tomosynthesis examinations.

Thermographic Profiles in Livestock Systems under Full Sun and Shaded Pastures during an Extreme Climate Event in the Eastern Amazon, Brazil: El Niño of 2023.

The El Niño represents a substantial threat to pastures, affecting the availability of water, forage and compromising the sustainability of grazing areas, especially in the northern region of Brazil. Therefore, the objective of this study was to characterize the thermographic profile of three production systems in the Eastern Amazon, Brazil. The study was conducted on a rural cattle farm in Mojuí dos Campos, Pará, Brazil, between August and November 2023. The experiment involved livestock production systems, including traditional, silvopastoral and integrated, with different conditions of shade and access to the bathing area. An infrared thermographic (IRT) camera was used, recording temperatures in different zones, such as areas with trees, pastures with forage and exposed pastures. The highest mean temperatures (p = 0.02) were observed in pastures with full sun from August to November. On the other hand, the lowest average temperatures were recorded in areas shaded by chestnut trees (Bertholletia excelsa). The highest temperature ranges were found in sunny areas and the lowest were recorded in shaded areas. The highest temperatures were observed in the pasture in full sun, while the lowest were recorded in areas shaded by chestnut trees (p < 0.0001). The interaction between the systems and treatments revealed significant temperature differences (p < 0.0001), with the native trees showing an average temperature of 35.9 °C, lower than the grasses and soil, which reached 61.2 °C. This research concludes that, under El Niño in the Eastern Amazon, areas shaded by Brazil nut trees had lower temperatures, demonstrating the effectiveness of shade. Native trees, compared to grasses and soil, showed the ability to create cooler environments, highlighting the positive influence on different species such as sheep, goats and cattle.

Inteligencia artificial. Aplicación en las imágenes cardiovasculares y la prevención cardiovascular / Artificial Intelligence in Cardiovascular Imaging and Cardiovascular Prevention

RESUMEN La inteligencia artificial (IA) está basada en programas computacionales que pueden imitar el pensamiento humano y automatizar algunos procesos. En el ámbito médico se está estudiando hace más de 50 años, pero en los últimos años el crecimiento ha sido exponencial. El campo de las imágenes cardiovasculares es particularmente atractivo para aplicarla, dado que, guiadas por IA, personas no expertas pueden adquirir imágenes completas, automatizar procesos y mediciones, orientar diagnósticos, detectar hallazgos no visibles al ojo humano, realizar diagnósticos oportunistas de afecciones no buscadas en el estudio índice pero evaluables a través de las imágenes disponibles, o identificar patrones de asociación dentro de una gran cantidad de datos como fuente de generación de hipótesis. En el campo de la prevención cardiovascular, la IA se ha aplicado en diferentes escenarios con fines diagnósticos, pronósticos y terapéuticos en el manejo de algunos factores de riesgo cardiovascular, como las dislipidemias o la hipertensión arterial. Si bien existen limitaciones con el uso de la IA tales como el costo, la accesibilidad y la compatibilidad de los programas, la validez externa de los resultados en determinadas poblaciones, o algunos aspectos éticos-legales (privacidad de los datos), esta tecnología está en crecimiento vertiginoso y posiblemente revolucione la práctica médica actual.

ABSTRACT Artificial intelligence (AI) is based on computer programs that imitate human thinking and automate certain processes. Artificial intelligence has been studied in the medical field for over 50 years, but in recent years, its growth has been exponential. The field of cardiovascular imaging is particularly attractive since AI can guide non-experts in image acquisition, automate processes and measurements, guide diagnoses, detect findings not visible to the human eye, make opportunistic diagnoses of unexpected conditions in the index test, or identify patterns of association within a large amount of data as a source of hypothesis generation. In the field of cardiovascular prevention, AI has been used for diagnostic, prognostic, and therapeutic purposes in managing cardiovascular risk factors such as dyslipidemia and hypertension. While there are limitations to the use of AI, such as cost, accessibility, compatibility of programs, external validity of results in certain populations, and ethical-legal aspects such as data privacy, this technology is rapidly growing and is likely to revolutionize current medical practice.

OralEpitheliumDB: A Dataset for Oral Epithelial Dysplasia Image Segmentation and Classification.

Early diagnosis of potentially malignant disorders, such as oral epithelial dysplasia, is the most reliable way to prevent oral cancer. Computational algorithms have been used as an auxiliary tool to aid specialists in this process. Usually, experiments are performed on private data, making it difficult to reproduce the results. There are several public datasets of histological images, but studies focused on oral dysplasia images use inaccessible datasets. This prevents the improvement of algorithms aimed at this lesion. This study introduces an annotated public dataset of oral epithelial dysplasia tissue images. The dataset includes 456 images acquired from 30 mouse tongues. The images were categorized among the lesion grades, with nuclear structures manually marked by a trained specialist and validated by a pathologist. Also, experiments were carried out in order to illustrate the potential of the proposed dataset in classification and segmentation processes commonly explored in the literature. Convolutional neural network (CNN) models for semantic and instance segmentation were employed on the images, which were pre-processed with stain normalization methods. Then, the segmented and non-segmented images were classified with CNN architectures and machine learning algorithms. The data obtained through these processes is available in the dataset. The segmentation stage showed the F1-score value of 0.83, obtained with the U-Net model using the ResNet-50 as a backbone. At the classification stage, the most expressive result was achieved with the Random Forest method, with an accuracy value of 94.22%. The results show that the segmentation contributed to the classification results, but studies are needed for the improvement of these stages of automated diagnosis. The original, gold standard, normalized, and segmented images are publicly available and may be used for the improvement of clinical applications of CAD methods on oral epithelial dysplasia tissue images.

A digital image-based colorimetric method for measuring free acidity in edible vegetable oils.

The standard method used to quantify free acidity (FA) in vegetable oil is neutralization titration, which requires many toxic chemicals and depends on an analyst's experience in detecting endpoints. Here, a digital image colorimetry (DIC) method using a smartphone camera was developed to measure FA in vegetable oils. A cupric acetate solution was used to produce the colorimetric reaction. The coloured solutions were imaged, and R values (from the RGB colour system) were calibrated against the respective FAs in the standards. The FA values of the samples were determined by standard addition calibration. These results were compared to measurements of FA obtained by the standard titrimetric method. An excellent correlation was obtained, with an R2 of 0.98 and a mean absolute error of 0.06%. The chemicals needed for analysis were reduced by approximately 90%. Thus, DIC is a less subjective and more economical method for determining FA in vegetable oils.

Integrating Dermoscopic Images into PACS Using DICOM and Modality Worklist.

Dermatology is one of the medical fields outside the radiology service that uses image acquisition and analysis in its daily medical practice, mostly through digital dermoscopy imaging modality. The acquisition, transfer, and storage of dermatology images has become an important issue to resolve. We aimed to describe our experience in integrating dermoscopic images into PACS using DICOM as a guide for the health informatics and dermatology community. During 2022 we integrated the video dermoscopy equipment through a strategic plan with an 8-step procedure. We used the DICOM standard with Modality Worklist and Storage commitment. Three systems were involved (video dermoscopy software, the EHR, and PACS). We identified critical steps and faced many challenges, such as the lack of a final model of DICOM standard for dermatology images.