BraNet: a mobil application for breast image classification based on deep learning algorithms.

Mobile health apps are widely used for breast cancer detection using artificial intelligence algorithms, providing radiologists with second opinions and reducing false diagnoses. This study aims to develop an open-source mobile app named "BraNet" for 2D breast imaging segmentation and classification using deep learning algorithms. During the phase off-line, an SNGAN model was previously trained for synthetic image generation, and subsequently, these images were used to pre-trained SAM and ResNet18 segmentation and classification models. During phase online, the BraNet app was developed using the react native framework, offering a modular deep-learning pipeline for mammography (DM) and ultrasound (US) breast imaging classification. This application operates on a client-server architecture and was implemented in Python for iOS and Android devices. Then, two diagnostic radiologists were given a reading test of 290 total original RoI images to assign the perceived breast tissue type. The reader's agreement was assessed using the kappa coefficient. The BraNet App Mobil exhibited the highest accuracy in benign and malignant US images (94.7%/93.6%) classification compared to DM during training I (80.9%/76.9%) and training II (73.7/72.3%). The information contrasts with radiological experts' accuracy, with DM classification being 29%, concerning US 70% for both readers, because they achieved a higher accuracy in US ROI classification than DM images. The kappa value indicates a fair agreement (0.3) for DM images and moderate agreement (0.4) for US images in both readers. It means that not only the amount of data is essential in training deep learning algorithms. Also, it is vital to consider the variety of abnormalities, especially in the mammography data, where several BI-RADS categories are present (microcalcifications, nodules, mass, asymmetry, and dense breasts) and can affect the API accuracy model.

OW-SLR: Overlapping Windows on Semi-Local Region for Image Super-Resolution.

There has been considerable progress in implicit neural representation to upscale an image to any arbitrary resolution. However, existing methods are based on defining a function to predict the Red, Green and Blue (RGB) value from just four specific loci. Relying on just four loci is insufficient as it leads to losing fine details from the neighboring region(s). We show that by taking into account the semi-local region leads to an improvement in performance. In this paper, we propose applying a new technique called Overlapping Windows on Semi-Local Region (OW-SLR) to an image to obtain any arbitrary resolution by taking the coordinates of the semi-local region around a point in the latent space. This extracted detail is used to predict the RGB value of a point. We illustrate the technique by applying the algorithm to the Optical Coherence Tomography-Angiography (OCT-A) images and show that it can upscale them to random resolution. This technique outperforms the existing state-of-the-art methods when applied to the OCT500 dataset. OW-SLR provides better results for classifying healthy and diseased retinal images such as diabetic retinopathy and normals from the given set of OCT-A images.

Preliminary Report on Optical Coherence Tomography Angiography Biomarkers in Non-Responders and Responders to Intravitreal Anti-VEGF Injection for Diabetic Macular Oedema.

PURPOSE: To identify optical coherence tomography angiography (OCTA) biomarkers in patients who were treated for diabetic macular oedema (DME) with intravitreal anti-vascular endothelial growth factor (VEGF) injections and compare the OCTA parameters between responders and non-responders. METHODS: A retrospective cohort study of 61 eyes with DME who received at least one intravitreal anti-VEGF injection was included between July 2017 and October 2020. The subjects underwent a comprehensive eye examination followed by an OCTA examination before and after intravitreal anti-VEGF injection. Demographic data, visual acuity, and OCTA parameters were documented, and further analysis was performed pre- and post-intravitreal anti-VEGF injection. RESULTS: Out of 61 eyes which underwent intravitreal anti-VEGF injection for diabetic macular oedema, 30 were responders (group 1) and 31 were non-responders (group 2). We found that the responders (group 1) had a higher vessel density in the outer ring that was statistically significant (p = 0.022), and higher perfusion density was noted in the outer ring (p = 0.012) and full ring (p = 0.044) at levels of the superficial capillary plexus (SCP). We also observed a lower vessel diameter index in the deep capillary plexus (DCP) in responders when compared to non-responders (p < 0.00). CONCLUSION: The evaluation of SCP in OCTA in addition to DCP can result in a better prediction of treatment response and early management in diabetic macular oedema.

Outcome of Ivermectin in Cancer Treatment: An Experience in Loja-Ecuador.

(1) Background: Cancer is one of the leading causes of death worldwide, and trends in cancer incidence and mortality are increasing over last years in Loja-Ecuador. Cancer treatment is expensive because of social and economic issues which force the patients to look for other alternatives. One such alternative treatment is ivermectin-based antiparasitic, which is commonly used in treating cattle. This paper analyzed ivermectin use as cancer treatment in the rural area of the Loja province and the medical opinion regarding the use of ivermectin in humans. (2) Methods: The study used a mixed methodology using different sampling techniques such as observation, surveys, and interviews. (3) Results: The main findings show that 19% of the participants diagnosed with cancer take medicines based on ivermectin as alternative therapy to the cancer control and treatment without leaving treatment such as chemotherapy, radiotherapy, or immunotherapy, while 81% use it to treat other diseases. (4) Conclusions: Finally, we identify that the interviewed not only use IVM as anticancer treatment, but it is also used as a treatment against other diseases. Although the participants' opinions indicate that they feel improvements in their health after the third dose, the specialist considers that there is no authorization to prescribe these alternative treatments. In addition, they confirmed that currently, there is no scientific knowledge about the application of these treatments in humans and they do not recommend their application. Thus, the anticancer mechanism of ivermectin remains to be further investigated; therefore, we consider that it is important to continue with this research by proposing a new stage to evaluate and determine the pharmacological action of this type of drug through an in vitro study in different cultures of cancer cells.

Detection and Classification of Knee Osteoarthritis.

Osteoarthritis (OA) affects nearly 240 million people worldwide. Knee OA is the most common type of arthritis, especially in older adults. Physicians measure the severity of knee OA according to the Kellgren and Lawrence (KL) scale through visual inspection of X-ray or MR images. We propose a semi-automatic CADx model based on Deep Siamese convolutional neural networks and a fine-tuned ResNet-34 to simultaneously detect OA lesions in the two knees according to the KL scale. The training was done using a public dataset, whereas the validations were performed with a private dataset. Some problems of the imbalanced dataset were solved using transfer learning. The model results average of the multi-class accuracy is 61%, presenting better performance results for classifying classes KL-0, KL-3, and KL-4 than KL-1 and KL-2. The classification results were compared and validated using the classification of experienced radiologists.

Frontiers in Retinal Image Processing.

Visual impairment is considered as a primary global challenge in the present era [...].

Relationship Between Posterior Vitreous Detachment and Macular Dimensions in Myopic Eyes.

Background: Posterior vitreous detachment (PVD) can result in serious pathologic events at the vitreoretinal interface. This study aims to assess the relationship between PVD, macular thickness (MT), and the foveal avascular zone (FAZ) in myopic eyes. Methods: This retrospective study evaluated 63 myopic subjects' data who were examined between January 1 and June 30, 2019. The myopes were grouped on a severity scale, namely mild, moderate, high, and very high using their spherical equivalent (SE). The PVD classification was based on OCT images. The status of the PVD and MT were evaluated with the Macular Cube 200×200 images, the FAZ with an OCTA Angioplex, and the dimensions were calculated using a customized algorithm. Results: The study population (114 eyes) had a median (range) age was 26.00 (22.00-28.00) years and the females constituted 62.3% of the dataset. In this population, 10 eyes had no PVD in any quadrant, and 73 eyes had incomplete PVD in all four quadrants. The inferior quadrant had the maximum rate of PVD occurrences and the nasal quadrant had the least number of occurrences. High myopic eyes exhibited significantly increased low foveal volume (p = <0.01). The inferior part of the para- and perimacular area showed a significant thinning in very high myopic eyes (p = <0.050). The very high myopic eyes showed a significant alteration of FAZ's circularity index (p = 0.002). Conclusion: In high and very high myopic eyes, an increasing trend of partial PVD is seen, most commonly in the inferior quadrant. A significant alteration in foveal volume and circularity index of the FAZ is seen in high and very high myopic eyes.

The [Formula: see text]-Wave of the Electroretinogram and Iron-Induced Oxidative Stress: A Model.

In photoreceptors of a dark adapted eye, the inward flux of sodium and calcium ions in the outer segment is balanced by the outward flux of potassium ions. But in the presence of light the creation of cyclic guanosine monophosphate in the outer segment decreases. Due to low concentration of cG (cyclic GMP) the channels in the outer segment open relatively less and thus the influx of calcium ion decreases, leading finally to hyperpolarization of the photoreceptors. We have analyzed theoretically the effect of oxidizing iron ions on the photoreceptors. In order to explain the effects of iron-induced oxidative stress, the different molecules and ions involved in phototransduction are quantified leading to a differential equation for calculating the electroretinogram a-wave voltage. The theoretical results are compared with published experimental data. In the presence of light, the iron ions could push outward the similarly charged calcium ions resulting in a small increase in the amount of inward calcium flux. Again, the presence of iron ions generates Reactive Oxygen Species, and ROS could attract the calcium ions which also increases the calcium flux. This will result in a reduction in the amplitude and slope of the a-wave voltage in the electroretinogram. These results are parametrized in terms of calcium ion concentrations. As the amplitude of the a-wave shows how much electrical signal is produced, its reduction indicates reduction in the visual signal. Thus, the increase in iron ions could explain the reduction in the electrical signal due to iron-induced oxidative stress.

ODTiD: Optic Nerve Head SD-OCT Image Dataset.

INTRODUCTION: Optic disc tilt (ODT) or tilted optic disc is a common finding in the general population. It is due to anomalous development caused by the malclosure of the embryonic optic fissure. ODT is commonly associated with high myopia as well as other conditions. In recent days, the common method to image the optic disc (OD) is by optical coherence tomography (OCT). To the best of our knowledge, there are no datasets of ODT available in the public domain. This dataset aims to make open access raw ODT OCT images to test out new image processing segmentation algorithms. METHODS: This dataset of ODT images contains both horizontal and vertical cross-sectional images obtained using spectral-domain optical coherence tomography (SD-OCT, Cirrus 5000, Carl Zeiss Meditec Inc., Dublin, CA). The optic disc cube 200×200 program was used and all the images are aligned with the center of the optic nerve head. This dataset includes images from both clinically normal (20 eyes) and myopic subjects (101 eyes). RESULTS: The dataset consists of clear (121) and manually marked (121) images resulting in a total of 242 images. The age distribution for all subjects combined is 27.24 ± 9.28 (range, 11.0-69.0) years. For normal subjects mean ± SD age distribution is 32.40 ± 17.23 years. Similarly, the myopia age distribution is 26.22 ± 6.37 years. Ground truth images, ie, manually segmented by a clinical expert are provided along with other meta-data includes age, gender, laterality, refractive error classification, spherical equivalent (SE), best-corrected visual acuity (BCVA), intraocular pressure (IOP), and axial length (AXL). CONCLUSION: This open, public database is online at the ICPSR website of the University of Michigan. The dataset can be used to test and validate newly developed automated segmentation algorithms.

Automated Detection and Diagnosis of Diabetic Retinopathy: A Comprehensive Survey.

Diabetic Retinopathy (DR) is a leading cause of vision loss in the world. In the past few years, artificial intelligence (AI) based approaches have been used to detect and grade DR. Early detection enables appropriate treatment and thus prevents vision loss. For this purpose, both fundus and optical coherence tomography (OCT) images are used to image the retina. Next, Deep-learning (DL)-/machine-learning (ML)-based approaches make it possible to extract features from the images and to detect the presence of DR, grade its severity and segment associated lesions. This review covers the literature dealing with AI approaches to DR such as ML and DL in classification and segmentation that have been published in the open literature within six years (2016-2021). In addition, a comprehensive list of available DR datasets is reported. This list was constructed using both the PICO (P-Patient, I-Intervention, C-Control, O-Outcome) and Preferred Reporting Items for Systematic Review and Meta-analysis (PRISMA) 2009 search strategies. We summarize a total of 114 published articles which conformed to the scope of the review. In addition, a list of 43 major datasets is presented.

Gelation Kinetics-Structure Analysis of pH-triggered Low Molecular Weight Hydrogelators.

Properties such as shear modulus, gelation time, structure of supramolecular hydrogels are strongly dependent on self-assembly, gelation triggering mechanism and processes used to form the gel. In our work we extend reported rheology analysis methodologies to pH-triggered supramolecular gels to understand structural insight using a model system based on N-N' Dibenzoyl-L-Cystine pH-triggered hydrogelator and Glucono-δ-Lactone as the trigger. We observed that Avrami growth model when applied to time-sweep rheological data of gels formed at lower trigger concentrations provide estimates of fractal dimension which agree well compared with visualization of the microstructure as seen via Confocal Laser Scanning Microscopy, for a range of gelator concentrations.

Rapid classification of glaucomatous fundus images.

We propose a new method for training convolutional neural networks (CNNs) and use it to classify glaucoma from fundus images. This method integrates reinforcement learning along with supervised learning and uses it for transfer learning. The training method uses hill climbing techniques via two different climber types, namely, "random movement" and "random detection," integrated with a supervised learning model through a stochastic gradient descent with momentum model. The model was trained and tested using the Drishti-GS and RIM-ONE-r2 datasets having glaucomatous and normal fundus images. The performance for prediction was tested by transfer learning on five CNN architectures, namely, GoogLeNet, DenseNet-201, NASNet, VGG-19, and Inception-Resnet v2. A five-fold classification was used for evaluating the performance, and high sensitivities while maintaining high accuracies were achieved. Of the models tested, the DenseNet-201 architecture performed the best in terms of sensitivity and area under the curve. This method of training allows transfer learning on small datasets and can be applied for tele-ophthalmology applications including training with local datasets.

Evaluation of Explainable Deep Learning Methods for Ophthalmic Diagnosis.

BACKGROUND: The lack of explanations for the decisions made by deep learning algorithms has hampered their acceptance by the clinical community despite highly accurate results on multiple problems. Attribution methods explaining deep learning models have been tested on medical imaging problems. The performance of various attribution methods has been compared for models trained on standard machine learning datasets but not on medical images. In this study, we performed a comparative analysis to determine the method with the best explanations for retinal OCT diagnosis. METHODS: A well-known deep learning model, Inception-v3 was trained to diagnose 3 retinal diseases - choroidal neovascularization (CNV), diabetic macular edema (DME), and drusen. The explanations from 13 different attribution methods were rated by a panel of 14 clinicians for clinical significance. Feedback was obtained from the clinicians regarding the current and future scope of such methods. RESULTS: An attribution method based on Taylor series expansion, called Deep Taylor, was rated the highest by clinicians with a median rating of 3.85/5. It was followed by Guided backpropagation (GBP), and SHapley Additive exPlanations (SHAP). CONCLUSION: Explanations from the top methods were able to highlight the structures for each disease - fluid accumulation for CNV, the boundaries of edema for DME, and bumpy areas of retinal pigment epithelium (RPE) for drusen. The most suitable method for a specific medical diagnosis task may be different from the one considered best for conventional tasks. Overall, there was a high degree of acceptance from the clinicians surveyed in the study.

FAZSeg: A New Software for Quantification of the Foveal Avascular Zone.

INTRODUCTION: Various ocular diseases and high myopia influence the anatomical reference point Foveal Avascular Zone (FAZ) dimensions. Therefore, it is important to segment and quantify the FAZs dimensions accurately. To the best of our knowledge, there is no automated tool or algorithms available to segment the FAZ's deep retinal layer. The paper describes a new open-access software with a Graphical User Interface (GUI) and compares the results with the ground truth (manual segmentation). METHODS: Ninety-three healthy normal subjects included 30 emmetropia and 63 myopic subjects without any sight-threatening retinal conditions, were included in the study. The 6mm x 6mm using the Angioplex protocol (Cirrus 5000 Carl Zeiss Meditec Inc., Dublin, CA) was used, and all the images were aligned with the centre of the fovea. Each FAZ image corresponding to dimensions 420×420 pixels were used in this study. These FAZ image dimensions for the superficial and deep layers were quantified using the New Automated Software Method (NAM). The NAM-based FAZ dimensions were validated with the ground truth. RESULTS: The age distribution for all 93 subjects was 28.02 ± 10.79 (range, 10.0-66.0) years. For normal subjects mean ± SD age distribution was 32.13 ± 16.27 years. Similarly, the myopia age distribution was 26.06 ± 6.06 years. The NAM had an accuracy of 91.40%. Moreover, the NAM on superficial layer FAZ gave a Dice Similarity Coefficient (DSC) score of 0.94 and Structural Similarity Index Metric (SSIM) of 0.97, while the NAM on deep layer FAZ gave a DSC score of 0.96 and SSIM of 0.98. CONCLUSION: A clinician-based GUI software was designed and tested on the FAZ images from deep and superficial layers. The NAM outperformed the device's inbuilt algorithm when measuring the superficial layer. This open-source software package is in the public domain and can be downloaded online.

The effect of refractive surgery on blur thresholds.

PURPOSE: The aim of this study was to measure blur thresholds before and after refractive surgery. METHODS: In this prospective cohort study conducted in a tertiary eye hospital in South India. Blur thresholds were measured for 30 young adult myopic patients 1 month prior to and after refractive surgery. Patients were asked to report three stages of blur, namely Detectable Blur (DB), Bothersome Blur (BB), and Non-resolvable Blur (NB). Blur was created by adding plus lenses (in steps of 0.12D) over their optimal subjective refraction. The blur judgments were made both monocularly and binocularly when looking through a 3 mm artificial pupil at one line above the best-corrected visual acuity. RESULTS: A total of 30 participants were included in this study (mean age = 25.5 ± 3.8 (20-36) years; 77% female). The mean binocular preoperative blur of this group was: DB = 0.39 ± 0.26D, BB = 0.74 ± 0.28D and NB = 1.04 ± 0.42D. The corresponding mean binocular blur one-month post-operatively was DB = 0.46 ± 0.28D, BB = 0.83 ± 0.35D, and NB = 1.21 ± 0.44D. Although there was a marginal increase in the blur thresholds postoperatively, the difference was not statistically significant (DB: P = 0.320; BB: P = 0.229; NB: P = 0.054). CONCLUSION: All three blur thresholds showed an insignificant minimal increase at 1 month post-operatively suggesting that patients adapt to the induced blur following refractive surgery. A longer follow up would reveal how the adaptation to blur would change with time.

Ophthalmic diagnosis using deep learning with fundus images - A critical review.

An overview of the applications of deep learning for ophthalmic diagnosis using retinal fundus images is presented. We describe various retinal image datasets that can be used for deep learning purposes. Applications of deep learning for segmentation of optic disk, optic cup, blood vessels as well as detection of lesions are reviewed. Recent deep learning models for classification of diseases such as age-related macular degeneration, glaucoma, and diabetic retinopathy are also discussed. Important critical insights and future research directions are given.

Explainable Deep Learning Models in Medical Image Analysis.

Deep learning methods have been very effective for a variety of medical diagnostic tasks and have even outperformed human experts on some of those. However, the black-box nature of the algorithms has restricted their clinical use. Recent explainability studies aim to show the features that influence the decision of a model the most. The majority of literature reviews of this area have focused on taxonomy, ethics, and the need for explanations. A review of the current applications of explainable deep learning for different medical imaging tasks is presented here. The various approaches, challenges for clinical deployment, and the areas requiring further research are discussed here from a practical standpoint of a deep learning researcher designing a system for the clinical end-users.

Locally pH controlled and directed growth of supramolecular gel microshapes using electrocatalytic nanoparticles.

Controlled localization of platinum nanoparticles (Pt NPs) at a solid support assisted by a polarized liquid-liquid interface is reported. Electrocatalytic water oxidation resulted in local pH modulation followed by the directed self-assembly of a dibenzoyl-l-cystine hydrogelator forming a structured hydrogel retaining the shape of the Pt NP deposit.

Is 25Hz enough to accurately measure a dynamic change in the ocular accommodation? / ¿Son suficientes 25Hz para medir con precisión un cambio dinámico en la acomodación ocular?

Background: Accommodation is often recorded at a low sampling rate using devices such as autorefractors that are designed to measure the static refractive error. It is therefore important to determine if that resolution is sufficient to accurately measure the dynamic properties of accommodation. The current study provides both theoretical and empirical evidence on the ideal sampling rate necessary to measure a dynamic response. Methods: Accommodative and disaccommodative step stimuli ranging from 1-3 D (1 D steps) were presented using a Badal optical system. Responses from 12 children (8-13 years) and 6 adults (20-35 years) were recorded using a dynamic photorefractor (DPR). Fast Fourier transformation was applied to the unsmoothed dynamic responses including position, velocity and acceleration. Also, velocity and acceleration main sequence (MS) characteristics were compared between three photorefractor conditions on 3 subjects. Results: The Nyquist sampling limit necessary to accurately estimate position, velocity and acceleration was at least 5, 10 and 70 Hz, respectively. Peak velocity and acceleration were significantly underestimated at a lower rate (p < 0.5). However, the slope of MS remained invariant with sampling rate (p > 0.5). Conclusion: Contrary to the previous findings, a dynamic accommodative response exhibited frequencies larger than 10Hz. Stimulus direction and amplitude had no influence on the frequencies present in the dynamic response. Peak velocity and acceleration can be significantly underestimated when sampled at a lower rate. Taken as a whole, low sampling rate instruments can accurately estimate static accommodation, however, caution needs to be exercised when using them for dynamic accommodation

La acomodación se registra a menudo a una tasa de muestreo baja, utilizando dispositivos tales como los autorrefractómetros que están diseñados para medir el error refractivo estático. Por tanto, es importante determinar si dicha resolución es suficiente para medir con precisión las propiedades dinámicas de la acomodación. El estudio actual aporta evidencia tanto teórica como empírica acerca de la tasa de muestreo necesaria para medir una respuesta dinámica. Métodos: Se presentaron estímulos de alteraciones de estimulación y relajación (desacomodación) de la acomodación que oscilaron entre 1 y 3 D (pasos de 1 D) utilizando un sistema óptico Badal. Se registraron las respuestas de 12 niños (de 8 a 13 años) y 6 adultos (de 20 a 35 años) utilizando un sistema de fotorrefracción dinámico (DPR). La transformación rápida de Fourier se aplicó a las respuestas dinámicas no uniformes incluyendo posición, velocidad y aceleración. También se compararon las características de la secuencia principal de velocidad y aceleración entre las tres situaciones del sistema de fotorrefracción en 3 sujetos. Resultados: El límite de muestreo de Nyquist necesario para calcular con precisión la posición, velocidad y aceleración fue de al menos 5, 10 y 70Hz respectivamente. La velocidad y aceleración máximas se subestimaron significativamente a una tasa inferior (p < 0,5). Sin embargo, la pendiente de la secuencia principal permaneció invariable con la tasa de muestreo (p > 0,5). Conclusión: Contrariamente a los hallazgos anteriores, la respuesta acomodativa dinámica mostró unas frecuencias superiores a 10Hz. La dirección y amplitud del estímulo no influyeron en las frecuencias presentes en la respuesta dinámica. La velocidad y aceleración máximas pueden subestimarse significativamente cuando se muestrean a una tasa menor. En conjunto, los instrumentos de baja tasa de muestreo pueden calcular con precisión la acomodación estática; sin embargo, debe actuarse con precaución a la hora de calcular la acomodación dinámica

Is 25Hz enough to accurately measure a dynamic change in the ocular accommodation?

BACKGROUND: Accommodation is often recorded at a low sampling rate using devices such as autorefractors that are designed to measure the static refractive error. It is therefore important to determine if that resolution is sufficient to accurately measure the dynamic properties of accommodation. The current study provides both theoretical and empirical evidence on the ideal sampling rate necessary to measure a dynamic response. METHODS: Accommodative and disaccommodative step stimuli ranging from 1-3D (1D steps) were presented using a Badal optical system. Responses from 12 children (8-13 years) and 6 adults (20-35 years) were recorded using a dynamic photorefractor (DPR). Fast Fourier transformation was applied to the unsmoothed dynamic responses including position, velocity and acceleration. Also, velocity and acceleration main sequence (MS) characteristics were compared between three photorefractor conditions on 3 subjects. RESULTS: The Nyquist sampling limit necessary to accurately estimate position, velocity and acceleration was at least 5, 10 and 70Hz, respectively. Peak velocity and acceleration were significantly underestimated at a lower rate (p<0.5). However, the slope of MS remained invariant with sampling rate (p>0.5). CONCLUSION: Contrary to the previous findings, a dynamic accommodative response exhibited frequencies larger than 10Hz. Stimulus direction and amplitude had no influence on the frequencies present in the dynamic response. Peak velocity and acceleration can be significantly underestimated when sampled at a lower rate. Taken as a whole, low sampling rate instruments can accurately estimate static accommodation, however, caution needs to be exercised when using them for dynamic accommodation.