Deep learning‐based automated detection of human knee joint's synovial fluid from magnetic resonance images with transfer learning

Shahzad, Ghazala; Rafiq, Nida; Mustafa, Ghulam; Ma, Jinwen

doi:10.1049/iet-ipr.2019.1646

Cited by 32 publications

(12 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Magnetic Resonance Imaging (MRI) is a medical imaging modality used to visualize the internal organs of the human body. The MRI is widely used for the diagnosis of a broad spectrum of diseases like ischemic stroke [1], Autism Spectrum Disorder (ASD) [2], Parkinson's disease [3], brain tumors [4], Schizophrenia [5], intracranial Tuberculosis [6], pancreatic cancer [7], Osteo Arthritis [8], prostate cancer [9] and Endometriosis [10]. Because of hardware limitations, images obtained from low-field MRI scanners are of low resolution, low acutance, and low contrast.…”

Section: Background and Problem Domainmentioning

confidence: 99%

A Post-Processing Algorithm for Boosting Contrast of MRI Images

Shiney¹,

Saleem²,

Rajinikanth³

et al. 2022

Computers, Materials &Amp; Continua

View full text Add to dashboard Cite

Low contrast of Magnetic Resonance (MR) images limits the visibility of subtle structures and adversely affects the outcome of both subjective and automated diagnosis. State-of-the-art contrast boosting techniques intolerably alter inherent features of MR images. Drastic changes in brightness features, induced by post-processing are not appreciated in medical imaging as the grey level values have certain diagnostic meanings. To overcome these issues this paper proposes an algorithm that enhance the contrast of MR images while preserving the underlying features as well. This method termed as Power-law and Logarithmic Modification-based Histogram Equalization (PLMHE) partitions the histogram of the image into two sub histograms after a power-law transformation and a log compression. After a modification intended for improving the dispersion of the sub-histograms and subsequent normalization, cumulative histograms are computed. Enhanced grey level values are computed from the resultant cumulative histograms. The performance of the PLMHE algorithm is compared with traditional histogram equalization based algorithms and it has been observed from the results that PLMHE can boost the image contrast without causing dynamic range compression, a significant change in mean brightness, and contrast-overshoot.

show abstract

Section: Background and Problem Domainmentioning

confidence: 99%

A Post-Processing Algorithm for Boosting Contrast of MRI Images

Shiney¹,

Saleem²,

Rajinikanth³

et al. 2022

Computers, Materials &Amp; Continua

View full text Add to dashboard Cite

show abstract

“…Recently, I3D proposed by Carreira et al [20], one of the state-of-the-art models, obtained inspiring accuracy with 98% on influential UCF-101 dataset. Among I3D, the two inflated inception-V1 (II-V1) sub-models are the core structure, which are constructed by repeating the convolution and pooling kernels of GoogLeNet [22]. I3D selected the weighted fusion strategy to make full use of the classification results from two II-V1 by weighting their outputs (video classification probabilities).…”

Section: Related Workmentioning

confidence: 99%

“…All the sub-models need massive training videos to effectively update their huge parameters. Considering that the number of collected multi-focus videos is limited and motivated by transfer learning technology [22,24], the training phases of the three sub-models all include two steps viz. pre-training and fine-tuning.…”

Section: Training Procedures Of Mi3dmentioning

confidence: 99%

Multi‐instance inflated 3D CNN for classifying urine red blood cells from multi‐focus videos

et al. 2022

IET Image Processing

View full text Add to dashboard Cite

Classifying urine red blood cells (U‐RBCs) is the core operation in diagnosing urinary system diseases (USDs). In this paper, based on a novel data type named multi‐focus video, a multi‐instance inflated 3D convolutional neural network (MI3D) is proposed. In order to accurately classifying U‐RBCs, the MI3D integrates inflated inception‐V1 with multi‐instance learning models. Compared with the existent U‐RBC classification methods relying on single focus images, the MI3D using multi‐focus videos effectively avoids the misclassification caused by the significant deformation of U‐RBCs with the focus of microscope changing. In addition, the MI3D can learn the typical shapes and deformation patterns of U‐RBCs from multi‐focal videos simultaneously. Therefore, the accuracy of MI3D exceeds the mainstream video classification models. There are totally 597 multi‐focus videos that include four types of U‐RBCs collected to verify the effectiveness of MI3D. Experimental results show that the classification accuracy of MI3D is inspiring with 94.4%, which is obviously higher than that of existed U‐RBC classification method (85.6%). The accuracy of MI3D also achieves the comparable level with the results by junior microscopist (95.6%). Lastly, the MI3D has powerful real‐time performance, whose classification speed reaches 1.4 times than that of the microscopist.

show abstract

“…With the above pre-processing, partitioning, and augmentation of the SCIAN and the HuSHeM datasets, we try to design a deep CNN architecture especially for the morphological classification of human sperm heads. The deep CNN architectures [23][24][25][26][27][28] obtained the top results in many complicated classification and regression tasks. Since the morphological classification of human sperm heads is an image classification task, it is proper to apply the deep CNN to solve such a complicated problem.…”

Section: Setmentioning

confidence: 99%

Deep Learning-Based Morphological Classification of Human Sperm Heads

Mustafa

2020

Diagnostics

Self Cite

View full text Add to dashboard Cite

Human infertility is considered as a serious disease of the reproductive system that affects more than 10% of couples across the globe and over 30% of the reported cases are related to men. The crucial step in the assessment of male infertility and subfertility is semen analysis that strongly depends on the sperm head morphology, i.e., the shape and size of the head of a spermatozoon. However, in medical diagnosis, the morphology of the sperm head is determined manually, and heavily depends on the expertise of the clinician. Moreover, this assessment as well as the morphological classification of human sperm heads are laborious and non-repeatable, and there is also a high degree of inter and intra-laboratory variability in the results. In order to overcome these problems, we propose a specialized convolutional neural network (CNN) architecture to accurately classify human sperm heads based on sperm images. It is carefully designed with several layers, and multiple filter sizes, but fewer filters and parameters to improve efficiency and effectiveness. It is demonstrated that our proposed architecture outperforms state-of-the-art methods, exhibiting 88% recall on the SCIAN dataset in the total agreement setting and 95% recall on the HuSHeM dataset for the classification of human sperm heads. Our proposed method shows the potential of deep learning to surpass embryologists in terms of reliability, throughput, and accuracy.

show abstract

Deep learning‐based automated detection of human knee joint's synovial fluid from magnetic resonance images with transfer learning

Cited by 32 publications

References 29 publications

A Post-Processing Algorithm for Boosting Contrast of MRI Images

A Post-Processing Algorithm for Boosting Contrast of MRI Images

Multi‐instance inflated 3D CNN for classifying urine red blood cells from multi‐focus videos

Deep Learning-Based Morphological Classification of Human Sperm Heads

Contact Info

Product

Resources

About