Beruflich Dokumente
Kultur Dokumente
Kiruthika P J a y a n th i K B
D ept. o f E C E D ept. o f ECE,
K.S. R angasam y College o f T echnology K.S. R angasam y College o f Technology
T am ilnadu, India T am ilnadu, India
e-m ail: kiruthi1012@ gm ail.com e-m ail: jayanthikb@ gm ail.com
Nirmala Madian
D ept. o f E C E
Sri Sakthi Institute o f E ngineering and Technology
Tam ilnadu, India
e-m ail: nirm alaooty@ gm ail.com
Abstract—Karyotyping of Banded Metaphase Chromosomes is im ages are tak en in this study as they are seen easily and
one of the preliminary steps used in cytogenetics to analyze the have unique light and dark bands.
chromosomes for diagnostic purposes. Deep learning is a G enerally there are 23 pairs o f chrom osom es in every
subfield of machine learning concerned with structure and hum an cell [1]. The first 22 pairs are autosom es and the 23rd
function of brain. It exploits a way to automate predictive p air is the sex chrom osom e. C hrom osom al abnorm alities are
analysis. The key aspect of deep learning is that the layers of related to the structure and num ber o f these 23 pairs o f
features are not designed by human engineers. They are chrom osom es. K aryotyping [2, 3] is a standard profile o f
learned from data using a general purpose learning procedure. chrom osom es as show n in Fig. 1.
This paper proposes a convolution based deep learning to
classify the chromosomes for automated karyotyping. The
developed architecture allows us to train and test images that
helps in predicting the chromosome abnormality. The
performance analysis is based on loss and accuracy curves and
the graphical representation clearly exhibits better
classification results for this architecture.
I. In t r o d u c t io n
111
size o f each filter being 3*3, and the activation function is to determ ine probabilistic confidence value o f output.
the rectifier function, w hich is used at fully connected layer
The last stage o f a convolutional neural netw ork is a they aggregate and recom bine features from the previous
classifier, called as dense layer. It needs individual features layer.
i.e., it needs a feature vector to perform classification. In
order to convert the output o f convolutional part o f the CN N III. Re s u l t s a nd D is c u s s io n s
into a 1D feature vector, flattening operation is done. It gets D iagnosis by deep learning are typically m ore objective
the output o f pooled im age pixels (2D array), flattens all its and accurate. To test the architecture, chrom osom e im ages
structure to create a one dim ensional single feature vector to are collected and total num ber o f chrom osom e im ages are
b e used by the dense layer fo r the final classification. A fully random ly divided into groups fo r training, validation and test
connected layer connects the set o f nodes got after the sets respectively. The training set consists o f 175 im ages and
flattening step. Fully connected layer perform s non-linear the testing set consists o f 83 im ages. E ach chrom osom e
transform ations o f the extracted features and classifies the im age is assigned a label from the 24 categories. F o r all the
inputs. This process com pletes the building up o f the experim ents, the resolution o f the im age is set to 64*64.
convolution neural netw ork model. Im age classification w ith D ata augm entation is done to increase the num ber o f data
CN N w orks quite w ell w hen enough training data is in the dataset. The netw ork is tested on various datasets
p rovided. w hich in turn is tested fo r different num ber o f epochs
Tw o m ajor categories o f im age classification techniques (iterations). H igher the learning rate, lesser num ber o f epochs
include unsupervised (calculated by softw are) and is required. L earning rate determ ines how quickly o r slowly
supervised (hum an-guided) classification. T raining im ages the update o f w eights have to takes place. F or sm aller values
are labeled in a supervised w ay by an analyst, b u t the feature o f learning rate, too m any iterations are needed to converge
learning and classification are autom atically done by to the b est values. Typically, learning rates are assigned at
softw are in an unsupervised way. W hile hum an visual im age random b ased on the user. A ll m odels w ere trained using
interpretation techniques rely o n shape, size, pattern, tone, convolutional neural networks. To regularize the netw ork
texture, shadows, and association, digital im age training, a sufficiently large num ber o f epochs are provided
interpretation relies m ainly o n color, i.e. o n com parisons o f w hile training each model. M odels are trained fo r different
digital num bers found in different bands in different parts o f values o f epochs. B y observing the validation results (i.e)
an im age. In deep-learning networks, each layer o f nodes loss and accuracy at each epoch, epoch can be varied to
trains o n a distinct set o f features b ased o n the previous obtain the highest validation accuracy. T his is im plem ented
lay er’s output. The further advancem ent into the neural net, using keras w ith tensorflow as backend.
the m ore com plex the features the nodes can recognize, since
112
(b), it can be seen that the validation accuracy becam e
slightly stable after 4-5 epochs and rarely increases at certain
epochs. In the beginning, validation accuracy linearly
increases w ith loss, then it does not increase much. The
training accuracy is high and training loss is quite low. B ut
the validation loss and accuracy are not better com pared to
training loss and accuracy, im plying that the m odel is over
fitting.
Epoch 25/25
100/190 - 5s 47ns/step • loss: 0.0714 • acc:
9725 • val.loss: 8.8124 • val.acc: l.M M
/hwe/researck/anaconda3/Ub/p)rtton3.6/slte -packapes/sktear r/cr »ss_»aUdat Ion.
:41: DeprecattoeMarnlng: This module »as deprecated in wrston 0.181* fa w o
the nodel.selectUn nodule into thick a ll the refactored classes and functions
re noved. also note that the interface of the ne» (V iterators are different f
n that of this nodule. This nodule udii be renewed tn 0.20.
'this nodule t i l l be renoved 1» 0.20.’ . Oeprecettoodarntnj)
predicted: fii.9954995U v___________________________________________
Figure 5. (b) output for Y chromosome
113
the 2012 IEEE Conference on Computer Vision and Pattern
Recognition (CVPR), 3642-3649, 2012
[2] Swati, Gaurav Gupta, Mohit Yadav, Monika Sharma, Lovekesh Vig,
Siamese Networks For Chromosome Classification, 2017 IEEE
conference on Computer Vision Workshops (ICCVW), 2017
[3] Altan, Novruz Allahverdi, Yakup Kutlu Diagnosis of Coronary
Artery Disease Using Deep Belief Networks, European journal of
engineering and natural sciences, Volume 2, Issue 1, pp. 29-36, 2017
[4] Wenzhong Yan, Lei Bai, Algorithms for Chromosome
Classification Engineering, 5, 400-403, 2013
[5] Xingwei Wang, Bin Zheng, Shibo Li, John J. Mulvihill, Marc C.
Wood, and Hong Liu, Automated Classification of Metaphase
Chromosomes: Optimization of an Adaptive Computerized Scheme,
Biomed Inform. 42(1): 22-31, 2009.
[6] D. Lu & Q. Weng ,A survey of image classification methods and
techniques for improving classification performance, International
Journal of Remote Sensing, 28, 5, 823-870, 2007
Figure 7(b). Plot between epoch and accuracy [7] Boaz Lerner, Toward A Completely Automatic Neural Network
Based Human Chromosome Analysis, IEEE Trans Syst 28(4):544-
M ostly, m any classification algorithm s w ork w ell w hen 552, 1998
sm all am ount o f data are available. A ccuracy o f classifiers [8] M. Zardoshti-Kermani and A. Afshordi, Classification of
Chromosomes Using Higher-Orde Neural Networks, Proceedings of
goes dow n w hen large dataset is taken fo r classification. B ut ICNN’95- International Conference on Neural Networks, IEEE, 1995.
deep learning perform s b etter even it com es fo r image [9] Sandeep Kumar, Zeeshan Khan, Anurag Jain, A Review of Content
classification w ith large am ount o f data. In the proposed Based Image Classification using Machine Learning Approach,
work, the chrom osom e prediction gives favorable results for International Journal of Advanced Computer Research, Volume-2
sex chrom osom es. F ew preprocessing steps like im age Number-3 Issue-5, 2012
[10] Walter P. Sweeney Jr., Mohamad T. Musavi,' and John N. Guidi,
straightening is adopted to im prove the perform ance. In
Classification of Chromosomes Using a Probabilistic Neural Network,
future work, deep learning convolutional neural netw ork can Journal of the international society for advancement of cytometry,
b e em ployed to classify all 22 pairs o f chrom osom es. Volume 16, issue1,Pages 17-24,1994
[11] Maximo E. Drets, Margery W. Shaw, “Specific banding patterns of
IV. Co n c l u s io n human chromosomes (heterochromatin/ Giemsa stain/ chromosome
bands)”, 1971, Proc. Nat. Acad. Sci. USA, Vol. 68, No. 9, pp. 2073
In this proposed w ork, classification o f chrom osom es is 2077
done using convolutional neural networks. The input [12] J.H. Tjio, A. Levan, The chromosome number in man, Hereditas 42
considered fo r CN N is the individual m etaphase (1956) 1-6.
chrom osom es. R ectified linear u n it (ReLU ) is the activation [13] William James Kenneth Cummino, Noi MaCumming Nevin, A
function used fo r extracting the features in CNN. The system for automated chromosome analysis, Humangenetik 7 (1969)
349-350.
extracted feature helps in the classification o f chrom osom es. [14] J. Graham, J. Piper, Automatic karyotype analysis, Humana, Totowa,
The proposed w ork gives an accuracy o f 100% fo r sex NJ, 1994.
chrom osm es b u t not as favourable fo r autosom es as the [15] S. B. Kotsiantis, Supervised Machine Learning: A Review of
chrom osom es do n ot have a proper size and structure. In the Classification Techniques, Informatica 31 (2007) 249-268
future w ork, a better deep learning algorithm can be [16] Chrysa Daiou, Alexandras Lambropoulos, Christoforos Markou,
Christos Maramis, Anastasios Delopoulos, Automatic Chromosome
em ployed to extract better features and the netw ork w ill be
Classification using Support Vector Machines
trained and tested w ith large num ber o f datasets for [17] T Arora, R Dhir, A review of metaphase chromosome image selection
im proved perform ance and accuracy. techniques for automatic karyotype generation Medical & biological
engineering & computing, Springer, 2016
Ac k n o w l ed g men t [18] F Abid, L Hamami, A survey of neural networks based automated
systems for human chromosome classification, Artificial Intelligence
T his project is supported by U niversity G rants Review, Springer, 2018.
Com m ission, India.
Re f e r ences
114