Your browser doesn't support javascript.
Show: 20 | 50 | 100
Results 1 - 20 de 48
Filter
1.
ACM Web Conference 2023 - Proceedings of the World Wide Web Conference, WWW 2023 ; : 3968-3977, 2023.
Article in English | Scopus | ID: covidwho-20244828

ABSTRACT

The COVID-19 pandemic has caused substantial damage to global health. Even though three years have passed, the world continues to struggle with the virus. Concerns are growing about the impact of COVID-19 on the mental health of infected individuals, who are more likely to experience depression, which can have long-lasting consequences for both the affected individuals and the world. Detection and intervention at an early stage can reduce the risk of depression in COVID-19 patients. In this paper, we investigated the relationship between COVID-19 infection and depression through social media analysis. Firstly, we managed a dataset of COVID-19 patients that contains information about their social media activity both before and after infection. Secondly, We conducted an extensive analysis of this dataset to investigate the characteristic of COVID-19 patients with a higher risk of depression. Thirdly, we proposed a deep neural network for early prediction of depression risk. This model considers daily mood swings as a psychiatric signal and incorporates textual and emotional characteristics via knowledge distillation. Experimental results demonstrate that our proposed framework outperforms baselines in detecting depression risk, with an AUROC of 0.9317 and an AUPRC of 0.8116. Our model has the potential to enable public health organizations to initiate prompt intervention with high-risk patients. © 2023 ACM.

2.
ACM International Conference Proceeding Series ; 2022.
Article in English | Scopus | ID: covidwho-20243125

ABSTRACT

Facial expression recognition (FER) algorithms work well in constrained environments with little or no occlusion of the face. However, real-world face occlusion is prevalent, most notably with the need to use a face mask in the current Covid-19 scenario. While there are works on the problem of occlusion in FER, little has been done before on the particular face mask scenario. Moreover, the few works in this area largely use synthetically created masked FER datasets. Motivated by these challenges posed by the pandemic to FER, we present a novel dataset, the Masked Student Dataset of Expressions or MSD-E, consisting of 1,960 real-world non-masked and masked facial expression images collected from 142 individuals. Along with the issue of obfuscated facial features, we illustrate how other subtler issues in masked FER are represented in our dataset. We then provide baseline results using ResNet-18, finding that its performance dips in the non-masked case when trained for FER in the presence of masks. To tackle this, we test two training paradigms: contrastive learning and knowledge distillation, and find that they increase the model's performance in the masked scenario while maintaining its non-masked performance. We further visualise our results using t-SNE plots and Grad-CAM, demonstrating that these paradigms capitalise on the limited features available in the masked scenario. Finally, we benchmark SOTA methods on MSD-E. The dataset is available at https://github.com/SridharSola/MSD-E. © 2022 ACM.

3.
Pattern Recognit ; 143: 109732, 2023 Nov.
Article in English | MEDLINE | ID: covidwho-20231102

ABSTRACT

Intelligent diagnosis has been widely studied in diagnosing novel corona virus disease (COVID-19). Existing deep models typically do not make full use of the global features such as large areas of ground glass opacities, and the local features such as local bronchiolectasis from the COVID-19 chest CT images, leading to unsatisfying recognition accuracy. To address this challenge, this paper proposes a novel method to diagnose COVID-19 using momentum contrast and knowledge distillation, termed MCT-KD. Our method takes advantage of Vision Transformer to design a momentum contrastive learning task to effectively extract global features from COVID-19 chest CT images. Moreover, in transfer and fine-tuning process, we integrate the locality of convolution into Vision Transformer via special knowledge distillation. These strategies enable the final Vision Transformer simultaneously focuses on global and local features from COVID-19 chest CT images. In addition, momentum contrastive learning is self-supervised learning, solving the problem that Vision Transformer is challenging to train on small datasets. Extensive experiments confirm the effectiveness of the proposed MCT-KD. In particular, our MCT-KD is able to achieve 87.43% and 96.94% accuracy on two publicly available datasets, respectively.

4.
Journal of Advanced Applied Scientific Research ; 4(4):50-60, 2022.
Article in English | Web of Science | ID: covidwho-2311094

ABSTRACT

Plants store a variety of important secondary metabolites with pharmacognostic and pharmacological implications, some of which have the potential to become super medicines in the future. In-vivo generation of these metabolites is influenced bya number of biotic and abiotic factors resulting in a constant accumulation of various phytochemicals and their derivatives that could be relevant in future medication research and development. There are over 70,000 plant species are employed ethnomedicinally in various ancient medical systems such as Ayurveda, Siddha, and Unani, as well as in Allopathy. The goal of this study is to look into the therapeutic potential of secondary metabolites as well as the probable pharmacological and pharmacognostic significance of the under-explored/underutilized plant Hyptissuaveolens (L.) Poit.Almost all parts of this plant are being employed in conventional drug to treat various diseases. It has been reported that it shows protection against peptic ulcer diseases and has anti-cancerous properties. The leaves of Hyptissuaveolenssecreted essential oil by hydrodistillation have been linked to the genus Hyptis' broad range of biological activity. It contains phytochemicals like alkaloids, tannins, saponins, flavonoids, terpenoids, minerals (like calcium, magnesium, sodium) and metals (like zinc and iron). The ursolic acid found in H. suaveolens can be used as a COVID-19 virus treatment agent. In addition, the ethnobotanical study claims that the beneficial plant has neuroprotective, antioxidant, antibacterial, antidiarrhoeal, anthelmintic, anti-inflammatory, wound healing, insecticidal, antimitotic, anti-proliferative, antisecretory, hepatoprotective, and acaricidal properties. The phytochemicals and extracts obtained from the plant have a great deal of therapeutic promise. As a result, we can use this plant for a variety of purposes.

5.
Soc Sci Med ; 327: 115799, 2023 06.
Article in English | MEDLINE | ID: covidwho-2308021

ABSTRACT

The nursing home sector was disproportionally affected by the COVID-19 pandemic and consequently, extreme mitigation strategies were taken in order to halt the spread of the virus. This research scrutinizes the manifestations of organizational trauma and healing amongst nursing home employees during the slow-burning pandemic. We aim to advance the contemporary debate around organizational healing that exclusively investigates fast-burning crises by translating these theories to a slow-burning crisis. Using participatory action research, we conducted two months of visual ethnographic fieldwork in a small-scale nursing home located in Amsterdam, the Netherlands from October to December 2021. Here, we present our findings constituting text and short videos according to the following four themes: (1) Emotional challenges in the workplace; (2) Cultural incompatibility of infection control strategies; (3) Navigating the ethics of decision-making; and (4) Organizational scars and healing perspectives. We propose the new concept of trauma distillation to describe and analyse how simmering organizational wounds are re-opened and purified to trigger a prolonged healing process in the context of slow-burning crises. Ultimately, this may lead to the acknowledgement and acceptance of such organizational wounds as multi-layered and intractable, aiming for a theoretical and empirical understanding of how to heal these. Our use of visual methods offers employees the opportunity to share their stories, make their suffering heard, and may contribute to nursing homes' processes of healing.


Subject(s)
COVID-19 , Humans , Pandemics , Nursing Homes , Anthropology, Cultural , Netherlands
6.
Electronics (Switzerland) ; 12(7), 2023.
Article in English | Scopus | ID: covidwho-2306047

ABSTRACT

A large number of mobile devices, smart wearable devices, and medical and health sensors continue to generate massive amounts of data, making edge devices' data explode and making it possible to implement data-driven artificial intelligence. However, the "data silos” and other issues still exist and need to be solved. Fortunately, federated learning (FL) can deal with "data silos” in the medical field, facilitating collaborative learning across multiple institutions without sharing local data and avoiding user concerns about data privacy. However, it encounters two main challenges in the medical field. One is statistical heterogeneity, also known as non-IID (non-independent and identically distributed) data, i.e., data being non-IID between clients, which leads to model drift. The second is limited labeling because labels are hard to obtain due to the high cost and expertise requirement. Most existing federated learning algorithms only allow for supervised training settings. In this work, we proposed a novel federated learning framework, MixFedGAN, to tackle the above issues in federated networks with dynamic aggregation and knowledge distillation. A dynamic aggregation scheme was designed to reduce the impact of current low-performing clients and improve stability. Knowledge distillation was introduced into the local generator model with a new distillation regularization loss function to prevent essential parameters of the global generator model from significantly changing. In addition, we considered two scenarios under this framework: complete annotated data and limited labeled data. An experimental analysis on four heterogeneous COVID-19 infection segmentation datasets and three heterogeneous prostate MRI segmentation datasets verified the effectiveness of the proposed federated learning method. © 2023 by the authors.

7.
Construction Management and Economics ; 41(5):402-427, 2023.
Article in English | ProQuest Central | ID: covidwho-2304999

ABSTRACT

The COVID-19 pandemic has been the largest global crisis in recent decades. Apart from the countless deaths and health emergencies, the pandemic has disrupted several industries—including construction. For example, a significant number of construction projects have been interrupted, delayed, and even abandoned. In such emergencies, information gathering and dissemination are vital for effective crisis management. The role of social media platforms such as YouTube, Facebook, and Twitter, as information sources, in these contexts has received much attention. The purpose of this investigation was to evaluate if YouTube can serve as a useful source of information for the construction industry in emergency situations—such as during the early stages of the COVID-19 pandemic. The assessment was undertaken by distilling the coverage of the COVID-19 pandemic as it relates to the construction industry from the content shared via YouTube by leveraging Latent Dirichlet Allocation (LDA) topic modelling. The investigation also compared the timeline with which relevant content was shared via YouTube and peer-reviewed research articles to make relative assessments. The findings suggest that YouTube offered significant and relevant coverage across six topics that include health and safety challenges, ongoing construction operation updates, workforce-related challenges, industry operations-related guidelines and advocacy, and others. Moreover, compared to the coverage of the COVID-19 pandemic in the research literature, YouTube offered more comprehensive and timely coverage of the pandemic as it relates to the construction industry. Accordingly, industry stakeholders may leverage YouTube as a valuable and largely untapped resource to aid in combating similar emergency situations.

8.
Healthcare (Basel) ; 11(7)2023 Apr 06.
Article in English | MEDLINE | ID: covidwho-2301940

ABSTRACT

Mental health problems are one of the various ills that afflict the world's population. Early diagnosis and medical care are public health problems addressed from various perspectives. Among the mental illnesses that most afflict the population is depression; its early diagnosis is vitally important, as it can trigger more severe illnesses, such as suicidal ideation. Due to the lack of homogeneity in current diagnostic tools, the community has focused on using AI tools for opportune diagnosis. Unfortunately, there is a lack of data that allows the use of IA tools for the Spanish language. Our work has a cross-lingual scheme to address this issue, allowing us to identify Spanish and English texts. The experiments demonstrated the methodology's effectiveness with an F1-score of 0.95. With this methodology, we propose a method to solve a classification problem for depression tweets (or short texts) by reusing English language databases with insufficient data to generate a classification model, such as in the Spanish language. We also validated the information obtained with public data to analyze the behavior of depression in Mexico during the COVID-19 pandemic. Our results show that the use of these methodologies can serve as support, not only in the diagnosis of depression, but also in the construction of different language databases that allow the creation of more efficient diagnostic tools.

9.
ACS Sustainable Chemistry and Engineering ; 11(5):1638-1642, 2023.
Article in English | Scopus | ID: covidwho-2271996

ABSTRACT

The COVID-19 pandemic has affected millions of people in the entire world and caused a shortage of several drugs, including propofol. Therefore, several protocols for propofol synthesis have been published in recent years. Herein, we present a process starting from paracetamol, a very common and abundant active pharmaceutical ingredient. Since the first three steps (Friedel-Crafts double alkylation, acetyl deprotection, and diazotization) are done in acidic media, a one-pot approach was developed. Furthermore, we observed that the extraction of the final product can be simplified by steam-distillation, leading to propofol in 47% isolated yield with high purity. This presented process could be an example of active pharmaceutical ingredient reuse since similar results were observed with commercial paracetamol tablets (with excipients) regardless of expiration date. © 2023 American Chemical Society.

10.
Connection Science ; 2023.
Article in English | Scopus | ID: covidwho-2268771

ABSTRACT

With the development of Medical Internet of Things (MIoT) technology and the global COVID-19 pandemic, hospitals gain access to patients' health data from remote wearable medical equipment. Federated learning (FL) addresses the difficulty of sharing data in remote medical systems. However, some key issues and challenges persist, such as heterogeneous health data stored in hospitals, which leads to high communication cost and low model accuracy. There are many approaches of federated distillation (FD) methods used to solve these problems, but FD is very vulnerable to poisoning attacks and requires a centralised server for aggregation, which is prone to single-node failure. To tackle this issue, we combine FD and blockchain to solve data sharing in remote medical system called FedRMD. FedRMD use reputation incentive to defend against poisoning attacks and store reputation values and soft labels of FD in Hyperledger Fabric. Experimenting on COVID-19 radiography and COVID-Chestxray datasets shows our method can reduce communication cost, and the performance is higher than FedAvg, FedDF, and FedGen. In addition, the reputation incentive can reduce the impact of poisoning attacks. © 2023 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.

11.
4th International Conference on Machine Learning for Cyber Security, ML4CS 2022 ; 13655 LNCS:501-515, 2023.
Article in English | Scopus | ID: covidwho-2268770

ABSTRACT

With the Internet of Things and medical technology development, patients use wearable telemedicine devices to transmit health data to hospitals. The need for data sharing for public health has become more urgent under the COVID-19 pandemic. Previously, security protection technology was difficult to solve the increasing security risks and challenges of telemedicine. To address the above hindrances, Federated learning (FL) solves the difficulty for companies and institutions to share user data securely. The global server iterative aggregates the model parameters from the local server instead of uploading the user's data directly to the cloud server. We propose a new model of federated distillation learning called FedTD, which allows the different models between local hospital servers and global servers. Unlike traditional federated learning, we combine the knowledge distillation method to solve the non-Independent Identically Distribution (non-IID) problem of patient medical data. It provides a security solution for sharing patients' medical information among hospitals. We tested our approach on the COVID-19 Radiography and COVID-Chestxray datasets to improve the model performance and reduce communication costs. Extensive experiments show that our FedTD significantly outperforms the state-of-the-art. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

12.
10th International Conference on Learning Representations, ICLR 2022 ; 2022.
Article in English | Scopus | ID: covidwho-2287080

ABSTRACT

We developed Distilled Graph Attention Policy Network (DGAPN), a reinforcement learning model to generate novel graph-structured chemical representations that optimize user-defined objectives by efficiently navigating a physically constrained domain. The framework is examined on the task of generating molecules that are designed to bind, noncovalently, to functional sites of SARS-CoV-2 proteins. We present a spatial Graph Attention (sGAT) mechanism that leverages self-attention over both node and edge attributes as well as encoding the spatial structure - this capability is of considerable interest in synthetic biology and drug discovery. An attentional policy network is introduced to learn the decision rules for a dynamic, fragment-based chemical environment, and state-of-the-art policy gradient techniques are employed to train the network with stability. Exploration is driven by the stochasticity of the action space design and the innovation reward bonuses learned and proposed by random network distillation. In experiments, our framework achieved outstanding results compared to state-of-the-art algorithms, while reducing the complexity of paths to chemical synthesis. © 2022 ICLR 2022 - 10th International Conference on Learning Representationss. All rights reserved.

13.
Intelligent Automation and Soft Computing ; 35(3):3517-3530, 2023.
Article in English | Scopus | ID: covidwho-2245735

ABSTRACT

The recent outbreak of the coronavirus disease of 2019 (Covid-19) has been causing many disruptions among the education systems worldwide, most of them due to the abrupt transition to online learning. The sudden upsurge in digital electronic devices usage, namely personal computers, laptops, tablets and smart-phones is unprecedented, which leads to a new wave of both mental and physical health problems among students, for example eye-related illnesses. The overexpo-sure to electronic devices, extended screen time usage and lack of outdoor sun-light have put a consequential strain on the student's ophthalmic health because of their young age and a relative lack of responsibility on their own health. Failure to take appropriate external measures to mitigate the negative effects of this process could lead to common ophthalmic illnesses such as myopia or more serious conditions. To remedy this situation, we propose a software solution that is able to track and capture images of its users' eyes to detect symptoms of eye illnesses while simultaneously giving them warnings and even offering treatments. To meet the requirements of a small and light model that is operable on low-end devices without information loss, we optimized the original MobileNetV2 model with depth-wise separable convolutions by altering the parameters in the last layers with an aim to minimize the resizing of the input image and obtained a new model which we call EyeNet. Combined with applying the knowledge distillation technique and ResNet-18 as a teacher model to train the student model, we have suc-cessfully increased the accuracy of the EyeNet model up to 87.16% and support the development of a model compatible with embedded systems with limited computing power, accessible to all students. © The Authors.

14.
Biofuels, Bioproducts and Biorefining ; 17(1):71-96, 2023.
Article in English | Scopus | ID: covidwho-2244630

ABSTRACT

In recent years, the production and consumption of fossil jet fuel have increased as a consequence of a rise in the number of passengers and goods transported by air. Despite the low demand caused by the coronavirus 2019 pandemic, an increase in the services offered by the sector is expected again. In an economic context still dependent on scarce oil, this represents a problem. There is also a problem arising from the fuel's environmental impact throughout its life cycle. Given this, a promising solution is the use of biojet fuel as renewable aviation fuel. In a circular economy framework, the use of lignocellulosic biomass in the form of sugar-rich crop residues allows the production of alcohols necessary to obtain biojet fuel. The tools provided by process intensification also make it possible to design a sustainable process with low environmental impact and capable of achieving energy savings. The goal of this work was to design an intensified process to produce biojet fuel from Mexican lignocellulosic biomass, with alcohols as intermediates. The process was modeled following a sequence of pretreatment/hydrolysis/fermentation/purification for the biomass-ethanol process, and dehydration/oligomerization/hydrogenation/distillation for ethanol-biojet process under the concept of distributed configuration. To obtain a cleaner, greener, and cheaper process, the purification zone of ethanol was intensified by employing a vapor side stream distillation column and a dividing wall column. Once designed, the entire process was optimized by employing the stochastic method of differential evolution with a tabu list to minimize the total annual cost and with the Eco-indicator-99 to evaluate the sustainability of the process. The results show that savings of 5.56% and a reduction of 1.72% in Eco-indicator-99 were achieved with a vapor side stream column in comparison with conventional distillation. On the other hand, with a dividing wall column, savings of 5.02% and reductions of 2.92% in Eco-indicator-99 were achieved. This process is capable of meeting a demand greater than 266 million liters of biojet fuel per year. However, the calculated sale price indicates that this biojet fuel still does not compete with conventional jet fuel produced in Mexico. © 2022 Society of Chemical Industry and John Wiley & Sons, Ltd. © 2022 Society of Chemical Industry and John Wiley & Sons, Ltd.

15.
Computer Vision, Eccv 2022, Pt Xxi ; 13681:627-643, 2022.
Article in English | Web of Science | ID: covidwho-2233939

ABSTRACT

As segmentation labels are scarce, extensive researches have been conducted to train segmentation networks with domain adaptation, semi-supervised or self-supervised learning techniques to utilize abundant unlabeled dataset. However, these approaches appear different from each other, so it is not clear how these approaches can be combined for better performance. Inspired by recent multi-domain image translation approaches, here we propose a novel segmentation framework using adaptive instance normalization (AdaIN), so that a single generator is trained to perform both domain adaptation and semi-supervised segmentation tasks via knowledge distillation by simply changing task-specific AdaIN codes. Specifically, our framework is designed to deal with difficult situations in chest X-ray radiograph (CXR) segmentation, where labels are only available for normal data, but the trained model should be applied to both normal and abnormal data. The proposed network demonstrates great generalizability under domain shift and achieves the state-of-the-art performance for abnormal CXR segmentation.

16.
ACS Sustainable Chemistry and Engineering ; 2022.
Article in English | Scopus | ID: covidwho-2233254

ABSTRACT

The COVID-19 pandemic has affected millions of people in the entire world and caused a shortage of several drugs, including propofol. Therefore, several protocols for propofol synthesis have been published in recent years. Herein, we present a process starting from paracetamol, a very common and abundant active pharmaceutical ingredient. Since the first three steps (Friedel-Crafts double alkylation, acetyl deprotection, and diazotization) are done in acidic media, a one-pot approach was developed. Furthermore, we observed that the extraction of the final product can be simplified by steam-distillation, leading to propofol in 47% isolated yield with high purity. This presented process could be an example of active pharmaceutical ingredient reuse since similar results were observed with commercial paracetamol tablets (with excipients) regardless of expiration date. © 2023 American Chemical Society.

17.
IAES International Journal of Artificial Intelligence ; 12(2):921-930, 2023.
Article in English | ProQuest Central | ID: covidwho-2230858

ABSTRACT

Facial expression recognition (FER) represents one of the most prevalent forms of interpersonal communication, which contains rich emotional information. But it became even more challenging during the times of COVID, where face masks became a mandatory protection measure, leading to the challenge of occluded lower-face during facial expression recognition. In this study, deep convolutional neural network (DCNN) represents the core of both our full-face FER system and our masked face FER model. The focus was on incorporating knowledge distillation in transfer learning between a teacher model, which is the full-face FER DCNN, and the student model, which is the masked face FER DCNN via the combination of both the loss from the teacher soft-labels vs the student soft labels and the loss from the dataset hard-labels vs the student hard-labels. The teacher-student architecture used FER2013 and a masked customized version of FER2013 as datasets to generate an accuracy of 69% and 61% respectively. Therefore, the study proves that the process of knowledge distillation may be used as a way for transfer learning and enhancing accuracy as a regular DCNN model (student only) would result in 46% accuracy compared to our approach (61% accuracy).

18.
Health Inf Sci Syst ; 11(1): 9, 2023 Dec.
Article in English | MEDLINE | ID: covidwho-2220290

ABSTRACT

3D medical image segmentation has an essential role in medical image analysis, while attention mechanism has improved the performance by a large margin. However, existing methods obtained the attention coefficient in a small receptive field, resulting in possible performance limitations. Radiologists usually scan all the slices first to have an overall idea of the target, and then analyze regions of interest in multiple 2D views in clinic practice. We simulate radiologists' recognition process and propose to exploit the 3D context information in a deeper manner for accurate 3D medical images segmentation. Due to the similarity of human body structure, medical images of different populations have highly similar shape and location information, so we use target region distillation to extract the common segmented region information. Particularly, we proposed two optimizations including Target Area Distillation and Section Attention. Target Area Distillation adds positions information to the original input to let the network has an initial attention of the target, while section attention performs attention extraction in three 2D sections thus with large range of receptive field. We compare our method against several popular networks in two public datasets including ImageCHD and COVID-19. Experimental results show that our proposed method improves the segmentation Dice score by 2-4% over the state-of-the-art methods. Our code has been released to the public (Anonymous link).

19.
IAES International Journal of Artificial Intelligence ; 12(2):921-930, 2023.
Article in English | Scopus | ID: covidwho-2203568

ABSTRACT

Facial expression recognition (FER) represents one of the most prevalent forms of interpersonal communication, which contains rich emotional information. But it became even more challenging during the times of COVID, where face masks became a mandatory protection measure, leading to the challenge of occluded lower-face during facial expression recognition. In this study, deep convolutional neural network (DCNN) represents the core of both our full-face FER system and our masked face FER model. The focus was on incorporating knowledge distillation in transfer learning between a teacher model, which is the full-face FER DCNN, and the student model, which is the masked face FER DCNN via the combination of both the loss from the teacher soft-labels vs the student soft labels and the loss from the dataset hard-labels vs the student hard-labels. The teacher-student architecture used FER2013 and a masked customized version of FER2013 as datasets to generate an accuracy of 69% and 61% respectively. Therefore, the study proves that the process of knowledge distillation may be used as a way for transfer learning and enhancing accuracy as a regular DCNN model (student only) would result in 46% accuracy compared to our approach (61% accuracy). © 2023, Institute of Advanced Engineering and Science. All rights reserved.

20.
2022 Ieee International Conference on Acoustics, Speech and Signal Processing (Icassp) ; : 3139-3143, 2022.
Article in English | Web of Science | ID: covidwho-2191818

ABSTRACT

Pothole classification has become an important task for road inspection vehicles to save drivers from potential car accidents and repair bills. Given the limited computational power and fixed number of training epochs, we propose iterative self knowledge distillation (ISKD) to train lightweight pothole classifiers. Designed to improve both the teacher and student models over time in knowledge distillation, ISKD outperforms the state-of-the-art self knowledge distillation method on three pothole classification datasets across four lightweight network architectures, which supports that self knowledge distillation should be done iteratively instead of just once. The accuracy relation between the teacher and student models shows that the student model can still benefit from a moderately trained teacher model. Implying that better teacher models generally produce better student models, our results justify the design of ISKD. In addition to pothole classification, we also demonstrate the efficacy of ISKD on six additional datasets associated with generic classification, fine-grained classification, and medical imaging application, which supports that ISKD can serve as a general-purpose performance booster without the need of a given teacher model and extra trainable parameters.

SELECTION OF CITATIONS
SEARCH DETAIL