Morphologic Classification and Automatic Diagnosis of Bacterial Vaginosis by Deep Neural Networks

AbstractBackgroundBacterial vaginosis (BV) was the most common condition for women’s health caused by the disruption of normal vaginal flora and an overgrowth of certain disease-causing bacteria, affecting 30-50% of women at some time in their lives. Gram stain followed by Nugent scoring (NS) based on bacterial morphotypes under the microscope was long considered golden standard for BV diagnosis. This conventional manual method was often considered labor intensive, time consuming, and variable results from person to person.MethodsWe developed four convolutional neural networks (CNN) models, and evaluated their ability to automatic identify vaginal bacteria and classify Nugent scores from microscope images. All the CNN models were first trained with 23280 microscopic images labeled with Nugent scores from top experts. A separate set of 5815 images were evaluated by the CNN models. The best CNN model was selected to generalize its application on an independent sets of 1082 images collecting from three teaching hospitals. Different hardwares were used to take images in hospitals.ResultsOur model could classify three Nugent Scores from images with high three classification accuracy of 89.3% (with 82.4% sensitivity and 96.6% specificity) on the 5815 test images, which was better diagnostic yield than the top-level technologists and obstetricians in China. The ability of generalization for our model was strong that it obtained 75.1%, which was 6.6% higher than the average of technologists.ConclusionThe CNN model over performed human healthcare practitioners on accuracy, efficiency and stability for BV diagnosis using microscopic image-based Nugent scores. The deep learning model may offer translational application in automating diagnosis of bacterial vaginosis with proper supporting hardware.

Download Full-text

Deep Neural Networks Offer Morphologic Classification and Diagnosis of Bacterial Vaginosis

Journal of Clinical Microbiology ◽

10.1128/jcm.02236-20 ◽

2020 ◽

Author(s):

Zhongxiao Wang ◽

Lei Zhang ◽

Min Zhao ◽

Ying Wang ◽

Huihui Bai ◽

...

Keyword(s):

Neural Networks ◽

Bacterial Vaginosis ◽

Deep Neural Networks ◽

Independent Test ◽

Golden Standard ◽

Test Sets ◽

Fully Connected ◽

Deep Learning Model ◽

Bacterial Morphotypes ◽

Better Than

Background: Bacterial vaginosis (BV) is caused by the excessive and imbalanced growth of bacteria in vagina, affecting 30-50% of women in their lives. Gram stain followed by Nugent scoring based on bacterial morphotypes under the microscope (NS) has been considered the golden standard for BV diagnosis, which is often labor-intensive, time-consuming, and variable results from person to person. Methods: We developed and optimized a convolutional neural networks (CNN) model, and evaluated its ability to automatically identify and classify three categories of Nugent scores from microscope images. The CNN model was first established with a panel of microscopic images with Nugent scores determined by experts. The model was trained by minimizing the cross entropy loss function and optimized by using a momentum optimizer. The separate test sets of images collected from three hospitals were evaluated by the CNN models. Results: The CNN model consisted of 25 convolutional layers, 2 pooling layers, and a fully connected layer. The model obtained 82.4% sensitivity and 96.6% specificity on the 5,815 validation images when considered altered vaginal flora and BV as the positive samples, which was better than the top-level technologists and obstetricians in China. The ability of generalization for our model was strong that it obtained 75.1% accuracy of three categories of Nugent scores on the independent test set of 1082 images, which was 6.6% higher than the average of three technologists, who are with a bachelor degree in medicine and eligible making diagnostic decisions. When three technologists ran one specimen in triplicate, the precision of three categories of Nugent scores was 54.0%. 103 samples diagnosed by two technologists at different days showed repeatability of 90.3%. Conclusion: The CNN model over-performed human healthcare practitioners on accuracy and stability for three categories of Nugent scores diagnosis. The deep learning model may offer translational applications in automating diagnosis of bacterial vaginosis with proper supporting hardware.

Download Full-text

Multimodal Multi-tasking for Skin Lesion Classification Using Deep Neural Networks

10.1007/978-3-030-90439-5_3 ◽

2021 ◽

pp. 27-38

Author(s):

Rafaela Carvalho ◽

João Pedrosa ◽

Tudor Nedelcu

Keyword(s):

Neural Networks ◽

Computer Vision ◽

Deep Learning ◽

Skin Lesion ◽

Deep Neural Networks ◽

Classification Performance ◽

Additional Information ◽

Deep Learning Model ◽

Lesion Classification

AbstractSkin cancer is one of the most common types of cancer and, with its increasing incidence, accurate early diagnosis is crucial to improve prognosis of patients. In the process of visual inspection, dermatologists follow specific dermoscopic algorithms and identify important features to provide a diagnosis. This process can be automated as such characteristics can be extracted by computer vision techniques. Although deep neural networks can extract useful features from digital images for skin lesion classification, performance can be improved by providing additional information. The extracted pseudo-features can be used as input (multimodal) or output (multi-tasking) to train a robust deep learning model. This work investigates the multimodal and multi-tasking techniques for more efficient training, given the single optimization of several related tasks in the latter, and generation of better diagnosis predictions. Additionally, the role of lesion segmentation is also studied. Results show that multi-tasking improves learning of beneficial features which lead to better predictions, and pseudo-features inspired by the ABCD rule provide readily available helpful information about the skin lesion.

Download Full-text

Zero-Keep Filter Pruning for Energy/Power Efficient Deep Neural Networks

Electronics ◽

10.3390/electronics10111238 ◽

2021 ◽

Vol 10 (11) ◽

pp. 1238

Author(s):

Yunhee Woo ◽

Dongyoung Kim ◽

Jaemin Jeong ◽

Young-Woong Ko ◽

Jeong-Gun Lee

Keyword(s):

Neural Networks ◽

Deep Learning ◽

High Performance ◽

Deep Neural Networks ◽

Learning Models ◽

Power Efficient ◽

Number Of Zeros ◽

Iot Devices ◽

Deep Learning Model ◽

Performance Computing

Recent deep learning models succeed in achieving high accuracy and fast inference time, but they require high-performance computing resources because they have a large number of parameters. However, not all systems have high-performance hardware. Sometimes, a deep learning model needs to be run on edge devices such as IoT devices or smartphones. On edge devices, however, limited computing resources are available and the amount of computation must be reduced to launch the deep learning models. Pruning is one of the well-known approaches for deriving light-weight models by eliminating weights, channels or filters. In this work, we propose “zero-keep filter pruning” for energy-efficient deep neural networks. The proposed method maximizes the number of zero elements in filters by replacing small values with zero and pruning the filter that has the lowest number of zeros. In the conventional approach, the filters that have the highest number of zeros are generally pruned. As a result, through this zero-keep filter pruning, we can have the filters that have many zeros in a model. We compared the results of the proposed method with the random filter pruning and proved that our method shows better performance with many fewer non-zero elements with a marginal drop in accuracy. Finally, we discuss a possible multiplier architecture, zero-skip multiplier circuit, which skips the multiplications with zero to accelerate and reduce energy consumption.

Download Full-text

Deep image captioning using an ensemble of CNN and LSTM based deep neural networks

Journal of Intelligent & Fuzzy Systems ◽

10.3233/jifs-189415 ◽

2020 ◽

pp. 1-9

Author(s):

Jafar A. Alzubi ◽

Rachna Jain ◽

Preeti Nagrath ◽

Suresh Satapathy ◽

Soham Taneja ◽

...

Keyword(s):

Neural Networks ◽

Deep Learning ◽

Deep Neural Networks ◽

Ensemble Model ◽

Image Captioning ◽

Deep Image ◽

The Given ◽

Deep Learning Model ◽

Image Caption Generation ◽

Image Caption

The paper is concerned with the problem of Image Caption Generation. The purpose of this paper is to create a deep learning model to generate captions for a given image by decoding the information available in the image. For this purpose, a custom ensemble model was used, which consisted of an Inception model and a 2-layer LSTM model, which were then concatenated and dense layers were added. The CNN part encodes the images and the LSTM part derives insights from the given captions. For comparative study, GRU and Bi-directional LSTM based models are also used for the caption generation to analyze and compare the results. For the training of images, the dataset used is the flickr8k dataset and for word embedding, dataset used is GloVe Embeddings to generate word vectors for each word in the sequence. After vectorization, Images are then fed into the trained model and inferred to create new auto-generated captions. Evaluation of the results was done using Bleu Scores. The Bleu-4 score obtained in the paper is 55.8%, and using LSTM, GRU, and Bi-directional LSTM respectively.

Download Full-text