Classification Accuracies of Malaria Infected Cells Using Deep - - PowerPoint PPT Presentation

Classification Accuracies of Malaria Infected Cells Using Deep Convolutional Neural Networks Based on Decompressed Images

Yuhang Dong, Zhuocheng Jiang, Hongda Shen, W. David Pan

• Dept. of Electrical & Computer Engineering, Univ. of Alabama in Huntsville (UAH)

• Problem Statement
• Machine Learning for Automated

Classification of Malaria Infected Cells

• Wholeslide Images
• Dataset of Cell Images for Malaria Infection
• Compression Methods
• Simulation Results and Case Study
• Conclusion

• In many biomedical applications, images are stored and

transmitted in the form of compressed images. However, typical pattern classifiers are trained using original images.

• There has been little prior study on how lossily decompressed

images would impact the classification performance.

• In a case study of automatic classification of malaria infected

cells, we used decompressed cell images as the inputs to deep convolutional neural networks.

• We evaluated how various lossy image compression

methods and varying compression ratios would impact the classification accuracies.

Problem Statement

• 214 million malaria cases, causing 438,000 death in 2015 (source:

WHO)

• In order to provide a reliable diagnosis, necessary training and

specialized human resource are required.

• Unfortunately, these resources are far from being adequate

and frequently often unavailable in underdeveloped areas where malaria has a marked predominance.

• Whole slide imaging (WSI), which scans the conventional glass

slides in order to produce high-resolution digital slides, is the most recent imaging modality being employed by pathology departments worldwide.

• WSI images allow for highly-accurate automated identification of

malaria infected cells.

Automated Identification of Malaria Infected Cells

• Classification accuracy of feature-based supervised

learning methods:

• 84% (SVM)
• 83.5% (Naïve Bayes Classifier)
• 85% (Three-layer Neural Network)
• Deep learning methods can extract a hierarchical

representation of the data from the input, with higher layers representing increasingly abstract concepts, which are increasingly invariant to transformations and scales.

• Study how lossy decompressed images would impact the classification

performance, by evaluating LeNet-5 (one CNN) on four methods:

• Bitplane reduction
• JPEG and JPEG 2000
• Sparse autoencoder

Red blood cell samples

Machine Learning for Malaria Detection

Wholeslide Images of Malaria Infection

Image of 258×258 with 100X magnification

Entire slide with cropped region delineated in green Whole Slide Image for malaria infected red blood cells from UAB However, there is NO publicly available high-resolution datasets for malaria to train and test deep neural networks – Need to build one!

Compilation of a Pathologist Curated Dataset

• Apply image morphological operations

to extract single cells.

• The dataset was curated by four UAB

pathologists.

• Each cell image was viewed and

labeled by at least two experienced pathologists from UAB Medical School (our collaborators).

• One cell image can only be considered

as infected and included in our final dataset if all the reviewers mark it positively, whereas it will be excluded otherwise.

• The same selection rule also applies

to the non-infected cells in our dataset.

• The dataset consists of 1,034 infected

cells and 1,531 non-infected cells. Link to the dataset

malaria (Data) label data Batch 64 Batch 100 Batch 64 Batch 100 scale (Power) scale conv1 (Convolution) kernel size: 5 stride: 1 pad: 0 conv1 pool1 (MAX Pooling) kernel size: 2 stride: 2 pad: 0 pool1 conv2 (Convolution) kernel size: 5 stride: 1 pad: 0 conv2 pool2 (MAX Pooling) kernel size: 2 stride: 2 pad: 0 pool2 ip1 (Inner Product) ip1 relu1 (ReLU) ip2 (Inner Product) 20 50 500 ip2 accuracy (Accuracy) loss (SoftmaxWithLoss) accuracy loss

Flowchart of automated malaria detection.

• 1. Convolutional layer
• 2. Pooling layer
• 3. ReLU layer
• 4. Fully connected layer
• 5. Loss layer

LeNet-5

LeNet-5 Convolutional Neural Network Architecture.

INPUT 60×60 C1: feature maps 20@56×56 S2: f. maps 20@28×28 C3: f. maps 50@24×24 S4: f. maps 50@12×12 C5: layer 500 F6: layer 500 OUTPUT 2 Convolutions Convolutions Subsampling Subsampling Full connection Full connection Gaussian connections

LeNet-5

• The image in the top row is the
• riginal malaria infected cell.
• Each of the four rows has eight

bitplane images, with the leftmost column representing the least significant bitplane (LSB) and the rightmost column the most significant bitplane (MSB).

• The second to forth row are

bitplane images from R, G and B channels, respectively.

• The bottom row are bitplanes

for combined RGB channels.

Bitplane Images

In this particular example, the MSB retain less features (e.g., the characteristic ring form of the parasite in an infected cell) than the second MSB, due to a majority of pixels in the original image have values above 128.

JPEG and JPEG 2000 Decompressed Image

• The number above an image is the corresponding compression ratio.
• The higher the compression ratio, the lower the quality of the reconstructed image.
• For JPEG, The ring form of the parasite is barely visible in the rightmost image,

which has the largest compression ratio.

• For JPEG-2000, The rightmost image has a compression ratio that more than

doubles that of co-located JPEG reconstructed image. Reconstructed image still retains the salient features of the original image such as the nucleus and ring form

• f the parasite.

JPEG JPEG-2K

Sparse Autoencoder

Compression Ratio = # of Pixels × 8 # of Neurons × 10

• Artificial neural network
• Unsupervised learning
• One encoder and one

decoder

• Use 10 neurons
• Training process: modify weight and bias to seek minimum difference between input

and reconstructed data.

• Loss in reconstructed data is inevitable, can be treated as lossy compression method.
• 8: number of bits per pixel
• 10: each neuron is a real number between 0 and 1, so 10 bits is enough to

represent number from 0 to 0.999.

Simulation Results

• Higher bitplane leads to

higher accuracy: flipped LSB only changes intensity by 1(20); flipped MSB changes intensity by 128(27)

• Bitplane from combined

RGB channels offers higher accuracy, albeit the cost of lower compression ratio.

• RGB component on

higher bits differ a lot leading to different accuracy. 1: keep only the least significant bitplane 8: keep only the most significant bitplane Classification accuracies of reconstructed bitplane images.

The classification accuracies of the reconstructed images using the JPEG and JPEG 2000 methods

Simulation Result and Discussions

• The higher the

compression ratio, the lower the visual quality, the lower the accuracy

• f classification.
• JPEG: higher than 95%

when ratio is 10

• JPEG-2K: maintain 95%

even with ratio of 30.

• Large reduction of

image size will be beneficial to storage and transmission

MNIST Dataset

• Left to right: combining the first 8 bitplanes (from MSB), goes down to

combining the first 1 bitplane (only MSB). Compression from 1:1 to 8:1.

• Top to bottom: Bitplane reduction, JPEG, JPEG 2000.

Simulation Result and Discussions

All accuracies are above 95% ! Reasons:

• High contrast between

digits and background

• Bitplane for background are

“1”, foreground are “0”.

• Even drop all lower seven

bitplanes, digits are still recognizable.

• certain degree of compres-

sion artifacts might indeed help the distinguishing features stand out better through deep learning Classification accuracies of the handwritten digits reconstructed from the compressed images in the MNIST datasets

Number of Neurons 100 70 50 30 20 15 10 Compression Ratio 6.27 8.96 12.54 20.90 31.36 41.80 62.70 Compression ratios VS number of neurons in single layer sparse autoencoder

Simulation Result and Discussions

• The higher the compression

ratio, the lower the visual quality, the lower the accuracy

• f classification.
• Even when compression higher

than 60, classifier can still achieve over 85% accuracy.

• Autoencoder offers much

wider range than other three method, while maintain reasonable good accuracy.

Conclusions

• Large reduction of medical image size would be very

beneficial to many telemedicine applications.

• we compared four compression methods: lossy

compression via bitplane reduction, JPEG and JPEG 2000, and sparse autoencoders.

• Bitplane reduction method had lower accuracy than JPEG

and JPEG 2000 methods.

• Autoencoders were capable of providing a much more

scalable compression ratios than the other three lossy compression methods, while maintaining a reasonably good classification accuracy.

• Further work: improve image quality of more “natural”

images like malaria cell samples using stacked autoencoders.