Evaluating large language models for annotating proteins.

In UniProtKB, up to date, there are more than 251 million proteins deposited. However, only 0.25% have been annotated with one of the more than 15000 possible Pfam family domains. The current annotation protocol integrates knowledge from manually curated family domains, obtained using sequence alignments and hidden Markov models. This approach has been successful for automatically growing the Pfam annotations, however at a low rate in comparison to protein discovery. Just a few years ago, deep learning models were proposed for automatic Pfam annotation. However, these models demand a considerable amount of training data, which can be a challenge with poorly populated families. To address this issue, we propose and evaluate here a novel protocol based on transfer learningThis requires the use of protein large language models (LLMs), trained with self-supervision on big unnanotated datasets in order to obtain sequence embeddings. Then, the embeddings can be used with supervised learning on a small and annotated dataset for a specialized task. In this protocol we have evaluated several cutting-edge protein LLMs together with machine learning architectures to improve the actual prediction of protein domain annotations. Results are significatively better than state-of-the-art for protein families classification, reducing the prediction error by an impressive 60% compared to standard methods. We explain how LLMs embeddings can be used for protein annotation in a concrete and easy way, and provide the pipeline in a github repo. Full source code and data are available at https://github.com/sinc-lab/llm4pfam.

MSDeepAMR: antimicrobial resistance prediction based on deep neural networks and transfer learning.

Introduction: Antimicrobial resistance (AMR) is a global health problem that requires early and effective treatments to prevent the indiscriminate use of antimicrobial drugs and the outcome of infections. Mass Spectrometry (MS), and more particularly MALDI-TOF, have been widely adopted by routine clinical microbiology laboratories to identify bacterial species and detect AMR. The analysis of AMR with deep learning is still recent, and most models depend on filters and preprocessing techniques manually applied on spectra. Methods: This study propose a deep neural network, MSDeepAMR, to learn from raw mass spectra to predict AMR. MSDeepAMR model was implemented for Escherichia coli, Klebsiella pneumoniae, and Staphylococcus aureus under different antibiotic resistance profiles. Additionally, a transfer learning test was performed to study the benefits of adapting the previously trained models to external data. Results: MSDeepAMR models showed a good classification performance to detect antibiotic resistance. The AUROC of the model was above 0.83 in most cases studied, improving the results of previous investigations by over 10%. The adapted models improved the AUROC by up to 20% when compared to a model trained only with external data. Discussion: This study demonstrate the potential of the MSDeepAMR model to predict antibiotic resistance and their use on external MS data. This allow the extrapolation of the MSDeepAMR model to de used in different laboratories that need to study AMR and do not have the capacity for an extensive sample collection.

Deep convolutional neural network for weld defect classification in radiographic images.

The quality of welds is critical to the safety of structures in construction, so early detection of irregularities is crucial. Advances in machine vision inspection technologies, such as deep learning models, have improved the detection of weld defects. This paper presents a new CNN model based on ResNet50 to classify four types of weld defects in radiographic images: crack, pore, non-penetration, and no defect. Stratified cross-validation, data augmentation, and regularization were used to improve generalization and avoid over-fitting. The model was tested on three datasets, RIAWELC, GDXray, and a private dataset of low image quality, obtaining an accuracy of 98.75 %, 90.255 %, and 75.83 %, respectively. The model proposed in this paper achieves high accuracies on different datasets and constitutes a valuable tool to improve the efficiency and effectiveness of quality control processes in the welding industry. Moreover, experimental tests show that the proposed approach performs well on even low-resolution images.

Task-based transferable deep-learning scatter correction in cone beam computed tomography: a simulation study.

Purpose: X-ray scatter significantly affects the image quality of cone beam computed tomography (CBCT). Although convolutional neural networks (CNNs) have shown promise in correcting x-ray scatter, their effectiveness is hindered by two main challenges: the necessity for extensive datasets and the uncertainty regarding model generalizability. This study introduces a task-based paradigm to overcome these obstacles, enhancing the application of CNNs in scatter correction. Approach: Using a CNN with U-net architecture, the proposed methodology employs a two-stage training process for scatter correction in CBCT scans. Initially, the CNN is pre-trained on approximately 4000 image pairs from geometric phantom projections, then fine-tuned using transfer learning (TL) on 250 image pairs of anthropomorphic projections, enabling task-specific adaptations with minimal data. 2D scatter ratio (SR) maps from projection data were considered as CNN targets, and such maps were used to perform the scatter prediction. The fine-tuning process for specific imaging tasks, like head and neck imaging, involved simulating scans of an anthropomorphic phantom and pre-processing the data for CNN retraining. Results: For the pre-training stage, it was observed that SR predictions were quite accurate (SSIM≥0.9). The accuracy of SR predictions was further improved after TL, with a relatively short retraining time (≈70 times faster than pre-training) and using considerably fewer samples compared to the pre-training dataset (≈12 times smaller). Conclusions: A fast and low-cost methodology to generate task-specific CNN for scatter correction in CBCT was developed. CNN models trained with the proposed methodology were successful to correct x-ray scatter in anthropomorphic structures, unknown to the network, for simulated data.

Development of a standardized methodology for transfer learning with QSAR models: a purely data-driven approach for source task selection.

Transfer learning is a machine learning technique that works well with chemical endpoints, with several papers confirming its efficiency. Although effective, because the choice of source/assistant tasks is non-trivial, the application of this technique is severely limited by the domain knowledge of the modeller. Considering this limitation, we developed a purely data-driven approach for source task selection that abstracts the need for domain knowledge. To achieve this, we created a supervised learning setting in which transfer outcome (positive/negative) is the variable to be predicted, and a set of six transferability metrics, calculated based on information from target and source datasets, are the features for prediction. We used the ChEMBL database to generate 100,000 transfers using random pairing, and with these transfers, we trained and evaluated our transferability prediction model (TP-Model). Our TP-Model achieved a 135-fold increase in precision while achieving a sensitivity of 92%, demonstrating a clear superiority against random search. In addition, we observed that transfer learning could provide considerable performance increases when applicable, with an average Matthews Correlation Coefficient (MCC) increase of 0.19 when using a single source and an average MCC increase of 0.44 when using multiple sources.

Reinforcement learning for in silico determination of adsorbate-substrate structures.

Reinforcement learning (RL) methods have helped to define the state of the art in the field of modern artificial intelligence, mostly after the breakthrough involving AlphaGo and the discovery of novel algorithms. In this work, we present a RL method, based on Q-learning, for the structural determination of adsorbate@substrate models in silico, where the minimization of the energy landscape resulting from adsorbate interactions with a substrate is made by actions on states (translations and rotations) chosen from an agent's policy. The proposed RL method is implemented in an early version of the reinforcement learning software for materials design and discovery (RLMaterial), developed in Python3.x. RLMaterial interfaces with deMon2k, DFTB+, ORCA, and Quantum Espresso codes to compute the adsorbate@substrate energies. The RL method was applied for the structural determination of (i) the amino acid glycine and (ii) 2-amino-acetaldehyde, both interacting with a boron nitride (BN) monolayer, (iii) host-guest interactions between phenylboronic acid and ß-cyclodextrin and (iv) ammonia on naphthalene. Density functional tight binding calculations were used to build the complex search surfaces with a reasonably low computational cost for systems (i)-(iii) and DFT for system (iv). Artificial neural network and gradient boosting regression techniques were employed to approximate the Q-matrix or Q-table for better decision making (policy) on next actions. Finally, we have developed a transfer-learning protocol within the RL framework that allows learning from one chemical system and transferring the experience to another, as well as from different DFT or DFTB levels.

Transfer learning for the generalization of artificial intelligence in breast cancer detection: a case-control study.

BACKGROUND: Some researchers have questioned whether artificial intelligence (AI) systems maintain their performance when used for women from populations not considered during the development of the system. PURPOSE: To evaluate the impact of transfer learning as a way of improving the generalization of AI systems in the detection of breast cancer. MATERIAL AND METHODS: This retrospective case-control Finnish study involved 191 women diagnosed with breast cancer and 191 matched healthy controls. We selected a state-of-the-art AI system for breast cancer detection trained using a large US dataset. The selected baseline system was evaluated in two experimental settings. First, we examined our private Finnish sample as an independent test set that had not been considered in the development of the system (unseen population). Second, the baseline system was retrained to attempt to improve its performance in the unseen population by means of transfer learning. To analyze performance, we used areas under the receiver operating characteristic curve (AUCs) with DeLong's test. RESULTS: Two versions of the baseline system were considered: ImageOnly and Heatmaps. The ImageOnly and Heatmaps versions yielded mean AUC values of 0.82±0.008 and 0.88±0.003 in the US dataset and 0.56 (95% CI=0.50-0.62) and 0.72 (95% CI=0.67-0.77) when evaluated in the unseen population, respectively. The retrained systems achieved AUC values of 0.61 (95% CI=0.55-0.66) and 0.69 (95% CI=0.64-0.75), respectively. There was no statistical difference between the baseline system and the retrained system. CONCLUSION: Transfer learning with a small study sample did not yield a significant improvement in the generalization of the system.

Towards a Corpus (and Language)-Independent Screening of Parkinson's Disease from Voice and Speech through Domain Adaptation.

End-to-end deep learning models have shown promising results for the automatic screening of Parkinson's disease by voice and speech. However, these models often suffer degradation in their performance when applied to scenarios involving multiple corpora. In addition, they also show corpus-dependent clusterings. These facts indicate a lack of generalisation or the presence of certain shortcuts in the decision, and also suggest the need for developing new corpus-independent models. In this respect, this work explores the use of domain adversarial training as a viable strategy to develop models that retain their discriminative capacity to detect Parkinson's disease across diverse datasets. The paper presents three deep learning architectures and their domain adversarial counterparts. The models were evaluated with sustained vowels and diadochokinetic recordings extracted from four corpora with different demographics, dialects or languages, and recording conditions. The results showed that the space distribution of the embedding features extracted by the domain adversarial networks exhibits a higher intra-class cohesion. This behaviour is supported by a decrease in the variability and inter-domain divergence computed within each class. The findings suggest that domain adversarial networks are able to learn the common characteristics present in Parkinsonian voice and speech, which are supposed to be corpus, and consequently, language independent. Overall, this effort provides evidence that domain adaptation techniques refine the existing end-to-end deep learning approaches for Parkinson's disease detection from voice and speech, achieving more generalizable models.

EvacuAI: An Analysis of Escape Routes in Indoor Environments with the Aid of Reinforcement Learning.

There is only a very short reaction time for people to find the best way out of a building in a fire outbreak. Software applications can be used to assist the rapid evacuation of people from the building; however, this is an arduous task, which requires an understanding of advanced technologies. Since well-known pathway algorithms (such as, Dijkstra, Bellman-Ford, and A*) can lead to serious performance problems, when it comes to multi-objective problems, we decided to make use of deep reinforcement learning techniques. A wide range of strategies including a random initialization of replay buffer and transfer learning were assessed in three projects involving schools of different sizes. The results showed the proposal was viable and that in most cases the performance of transfer learning was superior, enabling the learning agent to be trained in times shorter than 1 min, with 100% accuracy in the routes. In addition, the study raised challenges that had to be faced in the future.

A New Generative Model for Textual Descriptions of Medical Images Using Transformers Enhanced with Convolutional Neural Networks.

The automatic generation of descriptions for medical images has sparked increasing interest in the healthcare field due to its potential to assist professionals in the interpretation and analysis of clinical exams. This study explores the development and evaluation of a generalist generative model for medical images. Gaps were identified in the literature, such as the lack of studies that explore the performance of specific models for medical description generation and the need for objective evaluation of the quality of generated descriptions. Additionally, there is a lack of model generalization to different image modalities and medical conditions. To address these issues, a methodological strategy was adopted, combining natural language processing and features extraction from medical images and feeding them into a generative model based on neural networks. The goal was to achieve model generalization across various image modalities and medical conditions. The results showed promising outcomes in the generation of descriptions, with an accuracy of 0.7628 and a BLEU-1 score of 0.5387. However, the quality of the generated descriptions may still be limited, exhibiting semantic errors or lacking relevant details. These limitations could be attributed to the availability and representativeness of the data, as well as the techniques used.

Deep learning for Chilean native flora classification: a comparative analysis.

The limited availability of information on Chilean native flora has resulted in a lack of knowledge among the general public, and the classification of these plants poses challenges without extensive expertise. This study evaluates the performance of several Deep Learning (DL) models, namely InceptionV3, VGG19, ResNet152, and MobileNetV2, in classifying images representing Chilean native flora. The models are pre-trained on Imagenet. A dataset containing 500 images for each of the 10 classes of native flowers in Chile was curated, resulting in a total of 5000 images. The DL models were applied to this dataset, and their performance was compared based on accuracy and other relevant metrics. The findings highlight the potential of DL models to accurately classify images of Chilean native flora. The results contribute to enhancing the understanding of these plant species and fostering awareness among the general public. Further improvements and applications of DL in ecology and biodiversity research are discussed.

Gray-to-color image conversion in the classification of breast lesions on ultrasound using pre-trained deep neural networks.

Breast ultrasound (BUS) image classification in benign and malignant classes is often based on pre-trained convolutional neural networks (CNNs) to cope with small-sized training data. Nevertheless, BUS images are single-channel gray-level images, whereas pre-trained CNNs learned from color images with red, green, and blue (RGB) components. Thus, a gray-to-color conversion method is applied to fit the BUS image to the CNN's input layer size. This paper evaluates 13 gray-to-color conversion methods proposed in the literature that follow three strategies: replicating the gray-level image to all RGB channels, decomposing the image to enhance inherent information like the lesion's texture and morphology, and learning a matching layer. Besides, we introduce an image decomposition method based on the lesion's structural information to describe its inner and outer complexity. These gray-to-color conversion methods are evaluated under the same experimental framework using a pre-trained CNN architecture named ResNet-18 and a BUS dataset with more than 3000 images. In addition, the Matthews correlation coefficient (MCC), sensitivity (SEN), and specificity (SPE) measure the classification performance. The experimental results show that decomposition methods outperform replication and learning-based methods when using information from the lesion's binary mask (obtained from a segmentation method), reaching an MCC value greater than 0.70 and specificity up to 0.92, although the sensitivity is about 0.80. On the other hand, regarding the proposed method, the trade-off between sensitivity and specificity is better balanced, obtaining about 0.88 for both indices and an MCC of 0.73. This study contributes to the objective assessment of different gray-to-color conversion approaches in classifying breast lesions, revealing that mask-based decomposition methods improve classification performance. Besides, the proposed method based on structural information improves the sensitivity, obtaining more reliable classification results on malignant cases and potentially benefiting clinical practice.

Does Cognitive Impairment Impact Motor Learning? A Scoping Review of Elderly Individuals With Alzheimer's Disease and Mild Cognitive Impairment.

Individuals with cognitive impairment may have motor learning deficits due to the high engagement of cognitive mechanisms during motor skill acquisition. We conducted a scoping review to address the quality of current research on the relationship between cognitive impairments (i.e., deficits in attention, memory, planning and executive functions) and motor learning among older adults with Alzheimer's Disease or Mild Cognitive Impairment. After screening thousands of articles, we selected 15 studies describing cognitive assessment tools, experimental designs, and the severity of cognitive impairment. Although seven studies reported that cognitive impairment impaired motor learning, most studies included a high risk of bias. We identified multiple assessment tools across these studies that make comparisons among findings difficult. Future research in this area should focus on the influence of increased practice days during motor learning acquisition and incorporate both retention and transfer tests. Cognitive assessments should target the specific cognitive skills or deficits most closely related to the motor learning process.

Predicting socioeconomic indicators using transfer learning on imagery data: an application in Brazil.

Censuses and other surveys responsible for gathering socioeconomic data are expensive and time consuming. For this reason, in poor and developing countries there often is a long gap between these surveys, which hinders the appropriate formulation of public policies as well as the development of researches. One possible approach to overcome this challenge for some socioeconomic indicators is to use satellite imagery to estimate these variables, although it is not possible to replace demographic census surveys completely due to its territorial coverage, level of disaggregation of information and large set of information. Even though using orbital images properly requires, at least, a basic remote sensing knowledge level, these images have the advantage of being commonly free and easy to access. In this paper, we use daytime and nighttime satellite imagery and apply a transfer learning technique to estimate average income, GDP per capita and a constructed water index at the city level in two Brazilian states, Bahia and Rio Grande do Sul. The transfer learning approach could explain up to 64% of the variation in city-level variables depending on the state and variable. Although data from different countries may be considerably different, results are consistent with the literature and encouraging as it is a first analysis of its kind for Brazil.

An analysis of the influence of transfer learning when measuring the tortuosity of blood vessels.

BACKGROUND AND OBJECTIVE: Convolutional Neural Networks (CNNs) can provide excellent results regarding the segmentation of blood vessels. One important aspect of CNNs is that they can be trained on large amounts of data and then be made available, for instance, in image processing software. The pre-trained CNNs can then be easily applied in downstream blood vessel characterization tasks, such as the calculation of the length, tortuosity, or caliber of the blood vessels. Yet, it is still unclear if pre-trained CNNs can provide robust, unbiased, results in downstream tasks involving the morphological analysis of blood vessels. Here, we focus on measuring the tortuosity of blood vessels and investigate to which extent CNNs may provide biased tortuosity values even after fine-tuning the network to a new dataset under study. METHODS: We develop a procedure for quantifying the influence of CNN pre-training in downstream analyses involving the measurement of morphological properties of blood vessels. Using the methodology, we compare the performance of CNNs that were trained on images containing blood vessels having high tortuosity with CNNs that were trained on blood vessels with low tortuosity and fine-tuned on blood vessels with high tortuosity. The opposite situation is also investigated. RESULTS: We show that the tortuosity values obtained by a CNN trained from scratch on a dataset may not agree with those obtained by a fine-tuned network that was pre-trained on a dataset having different tortuosity statistics. In addition, we show that improving the segmentation accuracy does not necessarily lead to better tortuosity estimation. To mitigate the aforementioned issues, we propose the application of data augmentation techniques even in situations where they do not improve segmentation performance. For instance, we found that the application of elastic transformations was enough to prevent an underestimation of 8% of blood vessel tortuosity when applying CNNs to different datasets. CONCLUSIONS: The results highlight the importance of developing new methodologies for training CNNs with the specific goal of reducing the error of morphological measurements, as opposed to the traditional approach of using segmentation accuracy as a proxy metric for performance evaluation.

Retinal Lesion Segmentation Using Transfer Learning with an Encoder-Decoder CNN / Segmentación de Lesiones en la Retina Usando Transferencia de Conocimiento con CNN Encoder-Decoder

ABSTRACT Deep learning (DL) techniques achieve high performance in the detection of illnesses in retina images, but the majority of models are trained with different databases for solving one specific task. Consequently, there are currently no solutions that can be used for the detection/segmentation of a variety of illnesses in the retina in a single model. This research uses Transfer Learning (TL) to take advantage of previous knowledge generated during model training of illness detection to segment lesions with encoder-decoder Convolutional Neural Networks (CNN), where the encoders are classical models like VGG-16 and ResNet50 or variants with attention modules. This shows that it is possible to use a general methodology using a single fundus image database for the detection/segmentation of a variety of retinal diseases achieving state-of-the-art results. This model could be in practice more valuable since it can be trained with a more realistic database containing a broad spectrum of diseases to detect/segment illnesses without sacrificing performance. TL can help achieve fast convergence if the samples in the main task (Classification) and sub-tasks (Segmentation) are similar. If this requirement is not fulfilled, the parameters start from scratch.

RESUMEN Las técnicas de Deep Learning (DL) han demostrado un buen desempeño en la detección de anomalías en imágenes de retina, pero la mayoría de los modelos son entrenados en diferentes bases de datos para resolver una tarea en específico. Como consecuencia, actualmente no se cuenta con modelos que se puedan usar para la detección/segmentación de varias lesiones o anomalías con un solo modelo. En este artículo, se utiliza Transfer Learning (TL) con la cual se aprovecha el conocimiento adquirido para determinar si una imagen de retina tiene o no una lesión. Con este conocimiento se segmenta la imagen utilizando una red neuronal convolucional (CNN), donde los encoders o extractores de características son modelos clásicos como VGG-16 y ResNet50 o variantes con módulos de atención. Se demuestra así, que es posible utilizar una metodología general con bases de datos de retina para la detección/ segmentación de lesiones en la retina alcanzando resultados como los que se muestran en el estado del arte. Este modelo puede ser entrenado con bases de datos más reales que contengan una gama de enfermedades para detectar/ segmentar sin sacrificar rendimiento. TL puede ayudar a conseguir una convergencia rápida del modelo si la base de datos principal (Clasificación) se parece a la base de datos de las tareas secundarias (Segmentación), si esto no se cumple los parámetros básicamente comienzan a ajustarse desde cero.

Forecasting new diseases in low-data settings using transfer learning.

Recent infectious disease outbreaks, such as the COVID-19 pandemic and the Zika epidemic in Brazil, have demonstrated both the importance and difficulty of accurately forecasting novel infectious diseases. When new diseases first emerge, we have little knowledge of the transmission process, the level and duration of immunity to reinfection, or other parameters required to build realistic epidemiological models. Time series forecasts and machine learning, while less reliant on assumptions about the disease, require large amounts of data that are also not available in early stages of an outbreak. In this study, we examine how knowledge of related diseases can help make predictions of new diseases in data-scarce environments using transfer learning. We implement both an empirical and a synthetic approach. Using data from Brazil, we compare how well different machine learning models transfer knowledge between two different dataset pairs: case counts of (i) dengue and Zika, and (ii) influenza and COVID-19. In the synthetic analysis, we generate data with an SIR model using different transmission and recovery rates, and then compare the effectiveness of different transfer learning methods. We find that transfer learning offers the potential to improve predictions, even beyond a model based on data from the target disease, though the appropriate source disease must be chosen carefully. While imperfect, these models offer an additional input for decision makers for pandemic response.

Recognition of esophagitis in endoscopic images using transfer learning.

BACKGROUND: Esophagitis is an inflammatory and damaging process of the esophageal mucosa, which is confirmed by endoscopic visualization and may, in extreme cases, result in stenosis, fistulization and esophageal perforation. The use of deep learning (a field of artificial intelligence) techniques can be considered to determine the presence of esophageal lesions compatible with esophagitis. AIM: To develop, using transfer learning, a deep neural network model to recognize the presence of esophagitis in endoscopic images. METHODS: Endoscopic images of 1932 patients with a diagnosis of esophagitis and 1663 patients without any pathological diagnosis provenient from the KSAVIR and HyperKSAVIR datasets were splitted in training (80%) and test (20%) and used to develop and evaluate a binary deep learning classifier built using the DenseNet-201 architecture, a densely connected convolutional network, with weights pretrained on the ImageNet image set and fine-tuned during training. The classifier model performance was evaluated in the test set according to accuracy, sensitivity, specificity and area under the receiver operating characteristic curve (AUC). RESULTS: The model was trained using Adam optimizer with a learning rate of 0.0001 and applying binary cross entropy loss function. In the test set (n = 719), the classifier achieved 93.32% accuracy, 93.18% sensitivity, 93.46% specificity and a 0.96 AUC. Heatmaps for spatial predictive relevance in esophagitis endoscopic images from the test set were also plotted. In face of the obtained results, the use of dense convolutional neural networks with pretrained and fine-tuned weights proves to be a good strategy for predictive modeling for esophagitis recognition in endoscopic images. In addition, adopting the classification approach combined with the subsequent plotting of heat maps associated with the classificatory decision gives greater explainability to the model. CONCLUSION: It is opportune to raise new studies involving transfer learning for the analysis of endoscopic images, aiming to improve, validate and disseminate its use for clinical practice.

Transfer learning in proteins: evaluating novel protein learned representations for bioinformatics tasks.

A representation method is an algorithm that calculates numerical feature vectors for samples in a dataset. Such vectors, also known as embeddings, define a relatively low-dimensional space able to efficiently encode high-dimensional data. Very recently, many types of learned data representations based on machine learning have appeared and are being applied to several tasks in bioinformatics. In particular, protein representation learning methods integrate different types of protein information (sequence, domains, etc.), in supervised or unsupervised learning approaches, and provide embeddings of protein sequences that can be used for downstream tasks. One task that is of special interest is the automatic function prediction of the huge number of novel proteins that are being discovered nowadays and are still totally uncharacterized. However, despite its importance, up to date there is not a fair benchmark study of the predictive performance of existing proposals on the same large set of proteins and for very concrete and common bioinformatics tasks. Therefore, this lack of benchmark studies prevent the community from using adequate predictive methods for accelerating the functional characterization of proteins. In this study, we performed a detailed comparison of protein sequence representation learning methods, explaining each approach and comparing them with an experimental benchmark on several bioinformatics tasks: (i) determining protein sequence similarity in the embedding space; (ii) inferring protein domains and (iii) predicting ontology-based protein functions. We examine the advantages and disadvantages of each representation approach over the benchmark results. We hope the results and the discussion of this study can help the community to select the most adequate machine learning-based technique for protein representation according to the bioinformatics task at hand.

Deep transfer learning for the recognition of types of face masks as a core measure to prevent the transmission of COVID-19.

The use of face masks in public places has emerged as one of the most effective non-pharmaceutical measures to lower the spread of COVID-19 infection. This has led to the development of several detection systems for identifying people who do not wear a face mask. However, not all face masks or coverings are equally effective in preventing virus transmission or illness caused by viruses and therefore, it appears important for those systems to incorporate the ability to distinguish between the different types of face masks. This paper implements four pre-trained deep transfer learning models (NasNetMobile, MobileNetv2, ResNet101v2, and ResNet152v2) to classify images based on the type of face mask (KN95, N95, surgical and cloth) worn by people. Experimental results indicate that the deep residual networks (ResNet101v2 and ResNet152v2) provide the best performance with the highest accuracy and the lowest loss.