Bi-FPNFAS: Bi-Directional Feature Pyramid Network for Pixel-Wise Face Anti-Spoofing by Leveraging Fourier Spectra.

The emergence of biometric-based authentication using modern sensors on electronic devices has led to an escalated use of face recognition technologies. While these technologies may seem intriguing, they are accompanied by numerous implicit drawbacks. In this paper, we look into the problem of face anti-spoofing (FAS) on a frame level in an attempt to ameliorate the risks of face-spoofed attacks in biometric authentication processes. We employed a bi-directional feature pyramid network (BiFPN) that is used for convolutional multi-scaled feature extraction on the EfficientDet detection architecture, which is novel to the task of FAS. We further use these convolutional multi-scaled features in order to perform deep pixel-wise supervision. For all of our experiments, we performed evaluations across all major datasets and attained competitive results for the majority of the cases. Additionally, we showed that introducing an auxiliary self-supervision branch tasked with reconstructing the inputs in the frequency domain demonstrates an average classification error rate (ACER) of 2.92% on Protocol IV of the OULU-NPU dataset, which is significantly better than the currently available published works on pixel-wise face anti-spoofing. Moreover, following the procedures of prior works, we performed inter-dataset testing, which further consolidated the generalizability of the proposed models, as they showed optimum results across various sensors without any fine-tuning procedures.

The use of venous Doppler to predict adverse kidney events in a general ICU cohort.

BACKGROUND: Changes in Doppler flow patterns of hepatic veins (HV), portal vein (PV) and intra-renal veins (RV) reflect right atrial pressure and venous congestion; the feasibility of obtaining these assessments and the clinical relevance of the findings is unknown in a general ICU population. This study compares the morphology of HV, PV and RV waveform abnormalities in prediction of major adverse kidney events at 30 days (MAKE30) in critically ill patients. STUDY DESIGN AND METHODS: We conducted a prospective observational study enrolling adult patients within 24 h of admission to the ICU. Patients underwent an ultrasound evaluation of the HV, PV and RV. We compared the rate of MAKE-30 events in patients with and without venous flow abnormalities in the hepatic, portal and intra-renal veins. The HV was considered abnormal if S to D wave reversal was present. The PV was considered abnormal if the portal pulsatility index (PPI) was greater than 30%. We also examined PPI as a continuous variable to assess whether small changes in portal vein flow was a clinically important marker of venous congestion. RESULTS: From January 2019 to June 2019, we enrolled 114 patients. HV abnormalities demonstrate an odds ratio of 4.0 (95% CI 1.4-11.2). PV as a dichotomous outcome is associated with an increased odds ratio of MAKE-30 but fails to reach statistical significance (OR 2.3 95% CI 0.87-5.96), but when examined as a continuous variable it demonstrates an odds ratio of 1.03 (95% CI 1.00-1.06). RV Doppler flow abnormalities are not associated with an increase in the rate of MAKE-30 INTERPRETATION: Obtaining hepatic, portal and renal venous Doppler assessments in critically ill ICU patients are feasible. Abnormalities in hepatic and portal venous Doppler are associated with an increase in MAKE-30. Further research is needed to determine if venous Doppler assessments can be useful measures in assessing right-sided venous congestion in critically ill patients.

Correlation between tibial nail length and olecrenon to 5th metacarpal head measurement: An anthropometric study.

PURPOSE: During fracture fixation, the size of tibial nail is a vital factor affecting the outcomes and thus preoperative estimation of tibial nail length is very important. This study aims to find out whether "olecranon to 5th metacarpal head" (O-MH) measurement can be used to reliably predict the tibial nail length. METHODS: This was a cross sectional study involving 100 volunteers. Measurements were done and recorded by two observers on two separate occasions. Tibial nail length estimation measurement was done from highest point of tibial tuberosity to the tip of the medial malleolus (TT-MM). O-MH measurement was taken from tip of olecranon to the tip of 5th metacarpal head with wrist in neutral position and hand clenched. Statistical analysis was done to find out correlation between two measurements and influence of age, gender and body mass index on them. RESULTS: Paired t-test showed no systematic error between the readings. Intraclass correlation coefficient showed strong agreement in inter and intra observer settings. Strong correlation was found between the TT-MM & O-MH measurements using Pearson's correlation coefficient test (r = 0.966). Hierarchical regression analysis showed age, gender and BMI have no statistically significant bearings on these measurements and their correlations. CONCLUSION: O-MH measurement is a useful and accurate means of estimating tibial nail length preoperatively.

CosSIF: Cosine similarity-based image filtering to overcome low inter-class variation in synthetic medical image datasets.

Crafting effective deep learning models for medical image analysis is a complex task, particularly in cases where the medical image dataset lacks significant inter-class variation. This challenge is further aggravated when employing such datasets to generate synthetic images using generative adversarial networks (GANs), as the output of GANs heavily relies on the input data. In this research, we propose a novel filtering algorithm called Cosine Similarity-based Image Filtering (CosSIF). We leverage CosSIF to develop two distinct filtering methods: Filtering Before GAN Training (FBGT) and Filtering After GAN Training (FAGT). FBGT involves the removal of real images that exhibit similarities to images of other classes before utilizing them as the training dataset for a GAN. On the other hand, FAGT focuses on eliminating synthetic images with less discriminative features compared to real images used for training the GAN. The experimental results reveal that the utilization of either the FAGT or FBGT method reduces low inter-class variation in clinical image classification datasets and enables GANs to generate synthetic images with greater discriminative features. Moreover, modern transformer and convolutional-based models, trained with datasets that utilize these filtering methods, lead to less bias toward the majority class, more accurate predictions of samples in the minority class, and overall better generalization capabilities. Code and implementation details are available at: https://github.com/mominul-ssv/cossif.

Predicting life satisfaction using machine learning and explainable AI.

Life satisfaction is a crucial facet of human well-being. Hence, research on life satisfaction is incumbent for understanding how individuals experience their lives and influencing interventions targeted at enhancing mental health and well-being. Life satisfaction has traditionally been measured using analog, complicated, and frequently error-prone methods. These methods raise questions concerning validation and propagation. However, this study demonstrates the potential for machine learning algorithms to predict life satisfaction with a high accuracy of 93.80% and a 73.00% macro F1-score. The dataset comes from a government survey of 19000 people aged 16-64 years in Denmark. Using feature learning techniques, 27 significant questions for assessing contentment were extracted, making the study highly reproducible, simple, and easily interpretable. Furthermore, clinical and biomedical large language models (LLMs) were explored for predicting life satisfaction by converting tabular data into natural language sentences through mapping and adding meaningful counterparts, achieving an accuracy of 93.74% and macro F1-score of 73.21%. It was found that life satisfaction prediction is more closely related to the biomedical domain than the clinical domain. Ablation studies were also conducted to understand the impact of data resampling and feature selection techniques on model performance. Moreover, the correlation between primary determinants with different age brackets was analyzed, and it was found that health condition is the most important determinant across all ages. The best performing Machine Learning model trained in this study is deployed on a public server, ensuring unrestricted usage of the model. We highlight the advantages of machine learning methods for predicting life satisfaction and the significance of XAI for interpreting and validating these predictions. This study demonstrates how machine learning, large language models and XAI can jointly contribute to building trust and understanding in using AI to investigate human behavior, with significant ramifications for academics and professionals working to quantify and comprehend subjective well-being.

Experimental validation of computer-vision methods for the successful detection of endodontic treatment obturation and progression from noisy radiographs.

PURPOSE: (1) To evaluate the effects of denoising and data balancing on deep learning to detect endodontic treatment outcomes from radiographs. (2) To develop and train a deep-learning model and classifier to predict obturation quality from radiomics. METHODS: The study conformed to the STARD 2015 and MI-CLAIMS 2021 guidelines. 250 deidentified dental radiographs were collected and augmented to produce 2226 images. The dataset was classified according to endodontic treatment outcomes following a set of customized criteria. The dataset was denoised and balanced, and processed with YOLOv5s, YOLOv5x, and YOLOv7 models of real-time deep-learning computer vision. Diagnostic test parameters such as sensitivity (Sn), specificity (Sp), accuracy (Ac), precision, recall, mean average precision (mAP), and confidence were evaluated. RESULTS: Overall accuracy for all the deep-learning models was above 85%. Imbalanced datasets with noise removal led to YOLOv5x's prediction accuracy to drop to 72%, while balancing and noise removal led to all three models performing at over 95% accuracy. mAP saw an improvement from 52 to 92% following balancing and denoising. CONCLUSION: The current study of computer vision applied to radiomic datasets successfully classified endodontic treatment obturation and mishaps according to a custom progressive classification system and serves as a foundation to larger research on the subject matter.

Visual Diagnostics of Dental Caries through Deep Learning of Non-Standardised Photographs Using a Hybrid YOLO Ensemble and Transfer Learning Model.

BACKGROUND: Access to oral healthcare is not uniform globally, particularly in rural areas with limited resources, which limits the potential of automated diagnostics and advanced tele-dentistry applications. The use of digital caries detection and progression monitoring through photographic communication, is influenced by multiple variables that are difficult to standardize in such settings. The objective of this study was to develop a novel and cost-effective virtual computer vision AI system to predict dental cavitations from non-standardised photographs with reasonable clinical accuracy. METHODS: A set of 1703 augmented images was obtained from 233 de-identified teeth specimens. Images were acquired using a consumer smartphone, without any standardised apparatus applied. The study utilised state-of-the-art ensemble modeling, test-time augmentation, and transfer learning processes. The "you only look once" algorithm (YOLO) derivatives, v5s, v5m, v5l, and v5x, were independently evaluated, and an ensemble of the best results was augmented, and transfer learned with ResNet50, ResNet101, VGG16, AlexNet, and DenseNet. The outcomes were evaluated using precision, recall, and mean average precision (mAP). RESULTS: The YOLO model ensemble achieved a mean average precision (mAP) of 0.732, an accuracy of 0.789, and a recall of 0.701. When transferred to VGG16, the final model demonstrated a diagnostic accuracy of 86.96%, precision of 0.89, and recall of 0.88. This surpassed all other base methods of object detection from free-hand non-standardised smartphone photographs. CONCLUSION: A virtual computer vision AI system, blending a model ensemble, test-time augmentation, and transferred deep learning processes, was developed to predict dental cavitations from non-standardised photographs with reasonable clinical accuracy. This model can improve access to oral healthcare in rural areas with limited resources, and has the potential to aid in automated diagnostics and advanced tele-dentistry applications.

Mitigating carbon footprint for knowledge distillation based deep learning model compression.

Deep learning techniques have recently demonstrated remarkable success in numerous domains. Typically, the success of these deep learning models is measured in terms of performance metrics such as accuracy and mean average precision (mAP). Generally, a model's high performance is highly valued, but it frequently comes at the expense of substantial energy costs and carbon footprint emissions during the model building step. Massive emission of CO2 has a deleterious impact on life on earth in general and is a serious ethical concern that is largely ignored in deep learning research. In this article, we mainly focus on environmental costs and the means of mitigating carbon footprints in deep learning models, with a particular focus on models created using knowledge distillation (KD). Deep learning models typically contain a large number of parameters, resulting in a 'heavy' model. A heavy model scores high on performance metrics but is incompatible with mobile and edge computing devices. Model compression techniques such as knowledge distillation enable the creation of lightweight, deployable models for these low-resource devices. KD generates lighter models and typically performs with slightly less accuracy than the heavier teacher model (model accuracy by the teacher model on CIFAR 10, CIFAR 100, and TinyImageNet is 95.04%, 76.03%, and 63.39%; model accuracy by KD is 91.78%, 69.7%, and 60.49%). Although the distillation process makes models deployable on low-resource devices, they were found to consume an exorbitant amount of energy and have a substantial carbon footprint (15.8, 17.9, and 13.5 times more carbon compared to the corresponding teacher model). The enormous environmental cost is primarily attributable to the tuning of the hyperparameter, Temperature (τ). In this article, we propose measuring the environmental costs of deep learning work (in terms of GFLOPS in millions, energy consumption in kWh, and CO2 equivalent in grams). In order to create lightweight models with low environmental costs, we propose a straightforward yet effective method for selecting a hyperparameter (τ) using a stochastic approach for each training batch fed into the models. We applied knowledge distillation (including its data-free variant) to problems involving image classification and object detection. To evaluate the robustness of our method, we ran experiments on various datasets (CIFAR 10, CIFAR 100, Tiny ImageNet, and PASCAL VOC) and models (ResNet18, MobileNetV2, Wrn-40-2). Our novel approach reduces the environmental costs by a large margin by eliminating the requirement of expensive hyperparameter tuning without sacrificing performance. Empirical results on the CIFAR 10 dataset show that the stochastic technique achieves an accuracy of 91.67%, whereas tuning achieves an accuracy of 91.78%-however, the stochastic approach reduces the energy consumption and CO2 equivalent each by a factor of 19. Similar results have been obtained with CIFAR 100 and TinyImageNet dataset. This pattern is also observed in object detection classification on the PASCAL VOC dataset, where the tuning technique performs similarly to the stochastic technique, with a difference of 0.03% mAP favoring the stochastic technique while reducing the energy consumptions and CO2 emission each by a factor of 18.5.

Biological activities of Indian celery, Seseli diffusum (Roxb. ex Sm.) Sant. & Wagh.

In continuation of our work on Indian celery (Seseli diffusum (Roxb. ex Sm.) Santapau & Wagh; Umbelliferae), the fractionation of the 80% MeOH-H(2) O extract of the seeds was performed to identify the principles responsible for its folk use as an antispasmodic and diuretic. Several compounds were isolated as active components: seselin (1) and anthriscinol methyl ether (4) showed a selective cytotoxicity to some yeast strains. Compound 1 also showed spasmolytic activity. On the other hand, isopimpinellin (3) and isorutarin (5) exhibited a spasmogenic effect on the smooth muscle preparations. Compound 5 was also found to have antioxidant activity. Among them, compound 4 was isolated for the first time from this plant.

Knowledge distillation approach towards melanoma detection.

Melanoma is regarded as the most threatening among all skin cancers. There is a pressing need to build systems which can aid in the early detection of melanoma and enable timely treatment to patients. Recent methods are geared towards machine learning based systems where the task is posed as image recognition, tag dermoscopic images of skin lesions as melanoma or non-melanoma. Even though these methods show promising results in terms of accuracy, they are computationally quite expensive to train, that questions the ability of these models to be deployable in a clinical setting or memory constraint devices. To address this issue, we focus on building simple and performant models having few layers, less than ten compared to hundreds. As well as with fewer learnable parameters, 0.26 million (M) compared to 42.5 M using knowledge distillation with the goal to detect melanoma from dermoscopic images. First, we train a teacher model using a ResNet-50 to detect melanoma. Using the teacher model, we train the student model known as Distilled Student Network (DSNet) which has around 0.26 M parameters using knowledge distillation achieving an accuracy of 91.7%. We compare against ImageNet pre-trained models such MobileNet, VGG-16, Inception-V3, EfficientNet-B0, ResNet-50 and ResNet-101. We find that our approach works well in terms of inference runtime compared to other pre-trained models, 2.57 s compared to 14.55 s. We find that DSNet (0.26 M parameters), which is 15 times smaller, consistently performs better than EfficientNet-B0 (4 M parameters) in both melanoma and non-melanoma detection across Precision, Recall and F1 scores.

Late-onset capsular bag distention syndrome resistant to laser treatment: a case report.

Capsular bag distention syndrome (CBDS) or capsular block syndrome is a rare complication of cataract surgery. Neodymium-doped yttrium aluminum garnet (Nd:YAG) laser usually is effective treatment for CBDS. Rarely, surgical intervention is required in resistant cases (as in our case). Herein, we present the case of a 58-year-old male who presented to us with this condition.

Deep Learning Based Automatic Malaria Parasite Detection from Blood Smear and its Smartphone Based Application.

Malaria is a life-threatening disease that is spread by the Plasmodium parasites. It is detected by trained microscopists who analyze microscopic blood smear images. Modern deep learning techniques may be used to do this analysis automatically. The need for the trained personnel can be greatly reduced with the development of an automatic accurate and efficient model. In this article, we propose an entirely automated Convolutional Neural Network (CNN) based model for the diagnosis of malaria from the microscopic blood smear images. A variety of techniques including knowledge distillation, data augmentation, Autoencoder, feature extraction by a CNN model and classified by Support Vector Machine (SVM) or K-Nearest Neighbors (KNN) are performed under three training procedures named general training, distillation training and autoencoder training to optimize and improve the model accuracy and inference performance. Our deep learning-based model can detect malarial parasites from microscopic images with an accuracy of 99.23% while requiring just over 4600 floating point operations. For practical validation of model efficiency, we have deployed the miniaturized model in different mobile phones and a server-backed web application. Data gathered from these environments show that the model can be used to perform inference under 1 s per sample in both offline (mobile only) and online (web application) mode, thus engendering confidence that such models may be deployed for efficient practical inferential systems.

BanglaLekha-Isolated: A multi-purpose comprehensive dataset of Handwritten Bangla Isolated characters.

BanglaLekha-Isolated, a Bangla handwritten isolated character dataset is presented in this article. This dataset contains 84 different characters comprising of 50 Bangla basic characters, 10 Bangla numerals and 24 selected compound characters. 2000 handwriting samples for each of the 84 characters were collected, digitized and pre-processed. After discarding mistakes and scribbles, 1,66,105 handwritten character images were included in the final dataset. The dataset also includes labels indicating the age and the gender of the subjects from whom the samples were collected. This dataset could be used not only for optical handwriting recognition research but also to explore the influence of gender and age on handwriting. The dataset is publicly available at https://data.mendeley.com/datasets/hf6sf8zrkc/2.