Real-time deep learning-based image recognition for applications in automated positioning and injection of biological cells

Sadak, Ferhat; Saadat, Mozafar; Hajiyavand, Amir M.

doi:10.1016/j.compbiomed.2020.103976

Cited by 18 publications

(7 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…SVM was shown to be an effective technique for the diagnosis of COVID-19 [78]. Resnet50 is also an effective backbone that is being used for the development of neural networks in various biomedical applications [79]. However, the best performing pre-trained neural network models require a comprehensive evaluation as reported in [80] and [81] to diagnose COVID-19 using Xray images.…”

Section: State-ofthe-art In Deep Learning For Sars-cov-2 Diagnosismentioning

confidence: 99%

Electrochemical Biosensing and Deep Learning-Based Approaches in the Diagnosis of COVID-19: A Review

Sadak

Yıldırım

et al. 2022

IEEE Access

Self Cite

View full text Add to dashboard Cite

COVID-19 caused by the transmission of SARS-CoV-2 virus taking a huge toll on global health and caused life-threatening medical complications and elevated mortality rates, especially among older adults and people with existing morbidity. Current evidence suggests that the virus spreads primarily through respiratory droplets emitted by infected persons when breathing, coughing, sneezing, or speaking. These droplets can reach another person through their mouth, nose, or eyes, resulting in infection. The "gold standard" for clinical diagnosis of SARS-CoV-2 is the laboratory-based nucleic acid amplification test, which includes the reverse transcription-polymerase chain reaction (RT-PCR) test on nasopharyngeal swab samples. The main concerns with this type of test are the relatively high cost, long processing time, and considerable false-positive or false-negative results. Alternative approaches have been suggested to detect the SARS-CoV-2 virus so that those infected and the people they have been in contact with can be quickly isolated to break the transmission chains and hopefully, control the pandemic. These alternative approaches include electrochemical biosensing and deep learning. In this review, we discuss the current state-of-the-art technology used in both fields for public health surveillance of SARS-CoV-2 and present a comparison of both methods in terms of cost, sampling, timing, accuracy, instrument complexity, global accessibility, feasibility, and adaptability to mutations. Finally, we discuss the issues and potential future research approaches for detecting the SARS-CoV-2 virus utilizing electrochemical biosensing and deep learning. INDEX TERMSSARS-CoV-2; COVID-19; PCR; deep learning; electrochemical biosensor I. INTRODUCTION T He Director-General of the World Health Organization (WHO) declared the SARS-CoV-2 a pandemic on 12 March 2020 [1]. SARS-CoV-2 is extremely contagious and spreads similarly as SARS-CoV-1, by direct physical contact through respiratory droplets or touching of contaminated surfaces, as well as indirect contact by aerosolized SARS-CoV-2 [2]. Current evidence shows a significant proportion of those who are capable of transmitting the virus have no or barely visible symptoms while carrying the very high viral load [3]. This is noteworthy because asymptomatic COVID-19 patients are infectious and may act as silent drivers for the transmission of the virus. Pathogenic coronaviruses (CoVs) are a type of positive-

show abstract

Section: State-ofthe-art In Deep Learning For Sars-cov-2 Diagnosismentioning

confidence: 99%

Electrochemical Biosensing and Deep Learning-Based Approaches in the Diagnosis of COVID-19: A Review

Sadak

Yıldırım

et al. 2022

IEEE Access

Self Cite

View full text Add to dashboard Cite

show abstract

“…To further improve the performance of protein domain prediction, based on our originally developed DNN-Dom algorithm, we drew on the ideas of deep residual network and transfer learning in computer vision and natural language processing. The deep residual network was originally used in computer vision and then expanded to the field of computational biology such as protein structure prediction ( Wang et al , 2017 ), cell image recognition ( Sadak et al , 2020 ) and RNA secondary structure prediction ( Wang et al , 2021 ). Transfer learning can recognize and apply knowledge and skills learned in previous domains/tasks to novel domains/tasks and is particularly efficient when data are rich in a source domain but lacking in a target domain.…”

Section: Introductionmentioning

confidence: 99%

Res-Dom: predicting protein domain boundary from sequence using deep residual network and Bi-LSTM

Wang

Zhong

Xue

et al. 2022

Bioinformatics Advances

View full text Add to dashboard Cite

Motivation Protein domains are the basic units of proteins that can fold, function, and evolve independently. Protein domain boundary partition plays an important role in protein structure prediction, understanding their biological functions, annotating their evolutionary mechanisms, and protein design. Although there are many methods that have been developed to predict domain boundaries from protein sequence over the past two decades, there is still much room for improvement. Results In this article, a novel domain boundary prediction tool called Res-Dom was developed, which is based on deep residual network, Bi-LSTM and transfer learning. We used deep residual neural networks to extract higher-order residue-related information. In addition, we also used a pre-trained protein language model called ESM to extract sequence embedded features, which can summarize sequence context information more abundantly. To improve the global representation of these deep residual networks, a Bi-LSTM network was also designed to consider long-range interactions between residues. Res-Dom was then tested on an independent test set including 342 proteins and generated correct single-domain and multi-domain classifications with a Matthew’s correlation coefficient (MCC) of 0.668, which was 17.6% higher than the second-best compared method. For domain boundaries, the normalized domain overlapping score of Res-Dom was 0.849, which was 5% higher than the second-best compared method. Furthermore, Res-Dom required significantly less time than most of the recently developed state-of-the-art domain prediction methods. Availability all source code, datasets and model are available at http://isyslab.info/Res-Dom/.

show abstract

“…Deep learning (DL) approaches are currently widely recognized as the most promising technology for a wide range of computer vision applications, including medical robotics [11] and agricultural robotics [12], because they produce a superior performance to traditional machine learning techniques [13]. Faster-RCNN was used by Fanfang Gao et al to detect apples, with an average precision of 0.909, 0.899, 0.858, and 0.848 for non-occluded, leaf-occluded, branch/wire-occluded, and fruit-occluded fruit, respectively [14].…”

Section: Introductionmentioning

confidence: 99%

Strawberry Ripeness Assessment Via Camouflage-Based Data Augmentation for Automated Strawberry Picking Robot

Sadak

2022

Düzce Üniversitesi Bilim Ve Teknoloji Dergisi

Self Cite

View full text Add to dashboard Cite

Görüye dayalı çilek toplama ve yerleştirme, çilek hasat robotlarının görsel servo prosedürlerini doğru bir şekilde tamamlaması için ana hedeflerden biridir. Tarım robotları için çilek olgunluğunun saptanmasındaki ana zorluk oklüzyondur. Bu çalışmada, çilek hasat koşullarının doğal ortamını simüle etmek için kamuflaj tabanlı bir veri artırma stratejisi kullanılarak çilek olgunluğu tespiti önerilmiştir. Kamuflaj tabanlı veri artırma tekniğinin oklüzyon sorununun üstesinden gelmedeki etkisini bulmak için Yolov4, Yolov4 tiny ve Yolov4 scaled ve bunların geleneksel veri artırma ve kamuflaj tabanlı veri artırma türevleri kullanıldı. Daha sonra sonuçlar esas olarak mean Intersection over Union (IoU), F-1 score, average precision (AP), frame per second (fps) temel alınarak değerlendirildi. Kamuflaj tabanlı veri artırma tekniğine sahip Yolov4 tiny, ortalama 206 fps'de olgun ve olgunlaşmamış çilekler için %84 IoU doğruluğu ~%99 AP ile olgun ve olgunlaşmamış çilekleri tespit etmede üstün performans göstererek, tarım çilek hasat robotu operasyon ihtiyacını karşıladı. Önerilen tekniğin performansı, daha sonra, karmaşık ve oklüzyon olan bir çilek hasat ortamında performansını göstermek için bu çalışmada zorluk veri seti olarak adlandırılan bir veri seti kullanılarak başarılı bir şekilde test edildi. Kamuflaj tabanlı veri artırma tekniği, otonom çilek hasat robotları için olgun ve olgunlaşmamış çileğin tespit prosedürünü artırmaya yardımcı oldu.

show abstract

Real-time deep learning-based image recognition for applications in automated positioning and injection of biological cells

Cited by 18 publications

References 22 publications

Electrochemical Biosensing and Deep Learning-Based Approaches in the Diagnosis of COVID-19: A Review

Electrochemical Biosensing and Deep Learning-Based Approaches in the Diagnosis of COVID-19: A Review

Res-Dom: predicting protein domain boundary from sequence using deep residual network and Bi-LSTM

Strawberry Ripeness Assessment Via Camouflage-Based Data Augmentation for Automated Strawberry Picking Robot

Contact Info

Product

Resources

About