A comparative study on image-based snake identification using machine learning

An efficient densenet-based deep learning model for Big-4 snake species classification

Snakebite poses a significant health threat in numerous tropical and subtropical nations, with around 5.4 million cases reported annually, which results in 1.8-2.7 million instances of envenomation, underscoring its critical impact on public health. The 'BIG FOUR' group comprises the primary committers responsible for most snake bites in India. Effective management of snakebite victims is essential for prognosis, emphasizing the need for preventive measures to limit snakebite-related deaths. The proposed initiative seeks to develop a transfer learning-based image classification algorithm using DenseNet to identify venomous and non-venomous snakes automatically. The study comprehensively evaluates the image classification results, employing accuracy, F1-score, Recall, and Precision metrics. DenseNet emerges as a potent tool for multiclass snake image classification, achieving a notable accuracy rate of 86%. The proposed algorithm intends to be incorporated into an AI-based snake-trapping device with artificial prey made with tungsten wire and vibration motors to mimic heat and vibration signatures, enhancing its appeal to snakes. The proposed algorithm in this research holds promise as a primary tool for preventing snake bites globally, offering a path toward automated snake capture without human intervention. These findings are significant in preventing snake bites and advancing snakebite mitigation strategies.

Analyzing the biological traits of snakes in Avicenna's Canon of medicine and making a comparison with contemporary serpentology

Snakes are a group of reptiles in the order of squamata that have been the subject of scientific research for years. This study aimed to determine the biological characteristics of snakes that were mentioned in Avicenna's Canon of Medicine to compare with modern serpentology. Data were obtained from the Canon of Medicine using keywords concerning snakes and also appropriate literature in PubMed, Scopus, Web of Science, Scientific Information Database (SID), and IranDoc. Our results revealed that Avicenna divided snakes into three classes including highly, moderately, and slightly venomous snakes that are comparable to modern serpontology. Moreover, Avicenna clarified physiological factors such as age, gender, size, psychological state, hunger state, physical features, the type of living climate, habitat, and the time of snakebite. Considering the snake features presented in the Canon of Medicine, even though is not feasible to make a full comparison between Avicenna's snake knowledge and modern serpentology but some characteristics can be still applicable. Conclusively, the present study suggested that the criteria used for the identification and classification of snakes have changed from the medieval ages to the nowadays.

Medically important snakes and snakebite envenoming in Iran

Snakebite is a common health condition in Iran with a diverse snake fauna, especially in tropical southern and mountainous western areas of the country with plethora of snake species. The list of medically important snakes, circumstances and effects of their bite, and necessary medical care require critical appraisal and should be updated regularly. This study aims to review and map the distributions of medically important snake species of Iran, re-evaluate their taxonomy, review their venomics, describe the clinical effects of envenoming, and discuss medical management and treatment, including the use of antivenom. Nearly 350 published articles and 26 textbooks with information on venomous and mildly venomous snake species and snakebites of Iran, were reviewed, many in Persian (Farsi) language, making them relatively inaccessible to an international readership. This has resulted in a revised updated list of Iran's medically important snake species, with taxonomic revisions of some, compilation of their morphological features, remapping of their geographical distributions, and description of species-specific clinical effects of envenoming. Moreover, the antivenom manufactured in Iran is discussed, together with treatment protocols that have been developed for the hospital management of envenomed patients.

Animal Species Recognition with Deep Convolutional Neural Networks from Ecological Camera Trap Images

Accurate identification of animal species is necessary to understand biodiversity richness, monitor endangered species, and study the impact of climate change on species distribution within a specific region. Camera traps represent a passive monitoring technique that generates millions of ecological images. The vast numbers of images drive automated ecological analysis as essential, given that manual assessment of large datasets is laborious, time-consuming, and expensive. Deep learning networks have been advanced in the last few years to solve object and species identification tasks in the computer vision domain, providing state-of-the-art results. In our work, we trained and tested machine learning models to classify three animal groups (snakes, lizards, and toads) from camera trap images. We experimented with two pretrained models, VGG16 and ResNet50, and a self-trained convolutional neural network (CNN-1) with varying CNN layers and augmentation parameters. For multiclassification, CNN-1 achieved 72% accuracy, whereas VGG16 reached 87%, and ResNet50 attained 86% accuracy. These results demonstrate that the transfer learning approach outperforms the self-trained model performance. The models showed promising results in identifying species, especially those with challenging body sizes and vegetation.

Artificial intelligence-based snakebite identification using snake images, snakebite wound images, and other modalities of information: A systematic review

BACKGROUND AND OBJECTIVE

Artificial intelligence (AI) is widely applied in medical decision support systems. AI also plays an essential role in snakebite identification (SI). To date, no review has been conducted on AI-based SI. This work aims to identify, compare, and summarize the state-of-the-art AI methods in SI. Another objective is to analyze these methods and propose solutions for future directions.

METHODS

Searches were performed in PubMed, Web of Science, Engineering Village, and IEEE Xplore to identify the SI studies. The datasets, preprocessing, feature extraction, and classification algorithms of these studies were systematically reviewed. Then, their merits and defects were also analyzed and compared. Next, the quality of these studies was assessed by using the ChAIMAI checklist. Finally, solutions were proposed based on the limitations of current studies.

RESULTS

Twenty-six articles were included in the review. Traditional machine learning (ML) and deep learning (DL) algorithms were applied for the classification of snake images (Acc = 72 %∼98 %), wound images (Acc = 80 %∼100 %), and other modalities of information (Acc = 71.67 %∼97.6 %). According to the research quality assessment, one of the studies was considered to be of high quality. Most studies were flawed in data preparation, data understanding, validation, and deployment dimensions. In addition, we propose an active perception-based system framework for collecting images and bite forces and constructing a multi-modal dataset named "Digital Snake" to address the lack of high-quality datasets for DL algorithms to improve recognition accuracy and robustness. A Snakebite Identification, Treatment, and Management Assistive Platform architecture is also proposed as a decision support system for patients and doctors.

CONCLUSIONS

AI-based methods can quickly and accurately decide the snake species and classify venomous and non-venomous snakes. Current studies still have limitations in SI. Future studies based on AI methods should focus on constructing high-quality datasets and decision support systems for snakebite treatment.

Hierarchical cloud architecture for identifying the bite of "Egyptian cobra" based on deep learning and quantum particle swarm optimization

One of the most dangerous snake species is the "Egyptian cobra" which can kill a man in only 15 min. This paper uses deep learning techniques to identify the Egyptian cobra bite in an accurate manner based on an image of the marks of the bites. We build a dataset consisting of 500 images of cobra bites marks and 600 images of marks of other species of snakes that exist in Egypt. We utilize techniques such as multi-task learning, transfer learning and data augmentation to boost the generalization and accuracy of our model. We have achieved 90.9% of accuracy. We must keep the availability and accuracy of our model as much as possible. So, we utilize cloud and edge computing techniques to enhance the availability of our model. We have achieved 90.9% of accuracy, which is considered as an efficient result, not 100%, so it is normal for the system to perform sometimes wrong classifications. So, we suggest to re-train our model with the wrong predictions, whereas the edge computing units, where the classifier task is positioned, resend the wrong predictions to the cloud model, where the training process occurs, to retrain the model. This enhances the accuracy to the best level after a small period and increases the dataset size. We use the quantum particle swarm optimization technique to determine the optimal required number of edge nodes.

An artificial intelligence model to identify snakes from across the world: Opportunities and challenges for global health and herpetology

Background

Snakebite envenoming is a neglected tropical disease that kills an estimated 81,000 to 138,000 people and disables another 400,000 globally every year. The World Health Organization aims to halve this burden by 2030. To achieve this ambitious goal, we need to close the data gap in snake ecology and snakebite epidemiology and give healthcare providers up-to-date knowledge and access to better diagnostic tools. An essential first step is to improve the capacity to identify biting snakes taxonomically. The existence of AI-based identification tools for other animals offers an innovative opportunity to apply machine learning to snake identification and snakebite envenoming, a life-threatening situation.

Methodology

We developed an AI model based on Vision Transformer, a recent neural network architecture, and a comprehensive snake photo dataset of 386,006 training photos covering 198 venomous and 574 non-venomous snake species from 188 countries. We gathered photos from online biodiversity platforms (iNaturalist and HerpMapper) and a photo-sharing site (Flickr).

Principal findings

The model macro-averaged F1 score, which reflects the species-wise performance as averaging performance for each species, is 92.2%. The accuracy on a species and genus level is 96.0% and 99.0%, respectively. The average accuracy per country is 94.2%. The model accurately classifies selected venomous and non-venomous lookalike species from Southeast Asia and sub-Saharan Africa.

Conclusions

To our knowledge, this model’s taxonomic and geographic coverage and performance are unprecedented. This model could provide high-speed and low-cost snake identification to support snakebite victims and healthcare providers in low-resource settings, as well as zoologists, conservationists, and nature lovers from across the world.