Cervical Cancer Analysis by IRJET Journal

International Research Journal of Engineering and Technology (IRJET)

e-ISSN: 2395-0056

Volume: 09 Issue: 05 | May 2022

p-ISSN: 2395-0072

www.irjet.net

Cervical Cancer Analysis Aishwarya Kadam1, Divya Bharti B Kareti2, Priyanka Kurli3, Shreya Deshpande4, Pratibha Badiger5 1,2,3,4 Student,

Dept. of Information Science Engineering, SDMCET, Dharwad, Karnataka, India Dept. of Information Science Engineering, SDMCET, Dharwad, Karnataka, India ---------------------------------------------------------------------***--------------------------------------------------------------------and worldwide it is most feared disease. Due to abnormal Abstract – Gynecological cancers are among the most 5Professor,

common cancers in women and hence are an important public health issue. Due to the lack of cancer awareness programs, variable pathology, and dearth of proper screening facilities in developing countries such as India, most women report at advanced stages, adversely affecting the prognosis and clinical outcomes. Ovarian cancer has emerged as one of the most common malignancies affecting women in India. Cervical cancer remains the second most common cancer in women after breast cancer. The causes of cervical cancer are HPV (Human Papilloma Virus), smoking, oral contraceptives, multiple pregnancies. It can be prevented in adult women with early detection tests, such as the HPV test or PAP test, followed by treatment. Early detection tests can identify pre-cancerous lesions in the cervix, which can then be treated before the lesion develops into cervical cancer. In our proposed work, we used Machine Learning and Deep Learning algorithms to find a model capable of diagnosing cervical cancer with high accuracy and sensitivity.

Key Words: Machine Learning (ML), Convolution Neural Network (CNN), Deep Learning, Cervical Cancer.

1. INTRODUCTION Cervix is the lower, narrow end of the uterus that forms a canal between the uterus and vagina. Cervical cancer is caused by a sexually transmitted virus called Human Papilloma Virus (HPV) which accounts for 99.7% of all cervical cancer cases. In India, cervical cancer contributes to approximately 6–29% of all cancers in women and every year 122,844 women are diagnosed with cervical cancer and 67,477 die from this disease. In the early stages, there are no exact symptoms and side effects of the disease, but normal PAP smear screening is performed. Among other screening methods used to recognize malignant growth cells, the PAP smear test is excellent. The data collection of different types of cancer cells are taken and are preprocessed and trained to get a trained model which when used for the detection compares the accuracy of each class type in the dataset, the one with the highest class type accuracy is responded as a result.

Impact Factor value: 7.529

growth in the cervix cells, cervical cancer occurs and slowly it also spreads to the other organs of human body. Cervical cancer is caused by number reasons like human papillomavirus, using birth control pills, cigarette smoking, etc. In the initial stage, cervical cancer will not show any signs. However, if it is identified in earlier stage, it will be cured successfully. Nowadays, number of computer vision based approaches has been introduced to identify the cervical cancer disease and its stages. Cervical cancer arises due to uncontrolled development of cervical cells, they will not die instead they continue to divide. Literature reports that HPV virus, smoking, and weak immune system, etc are the causes of cervical cancer. Nowadays, death rate due to the cervical cancer is reduced significantly by detecting the cervical cancer in its early stage using the pap smear test. Screening process undergone for cervical cancer manually has higher issues of producing false negative rates in Pap smear test. Hence an alternate method came into existence called automated computer based technique to increase accuracy for testing cervical smears. Image processing techniques are proposed here to detect the cervical cancer early [2] as Cervical cancer is most common malignancy in female. It arising from the cervix. There are different treatment like primary surgery, primary radio therapy, chemo therapy and combination therapy. There are many techniques to diagnosis cervical cancer the important ones are Pap smear test, LBC test, HPV test, Biopsy and different screening techniques. The automatic screening of the cervical malignant growth cells has been created by means of morphological image processing techniques. Cervical most cancers are screened manually with the aid of the usage of the Pap smear test and LCB check which does not deliver correct classification effects in classifying the normal and uncommon cervical cells inside the cervix region of the uterus. The manually screened technique suffers from excessive faux fee because of human errors and also value effective to be executed by means of the usage of the professional cytologist. in this paper, several methods are proposed for the automatic detection of cervical cancer the usage of image processing techniques. The automatic techniques are achieved to supply correct outcomes and to make effective type of ordinary and atypical cells. The paper [3] proposed here is Cervical Cancer Diagnosis using CervixNet - A Deep Learning Approach. Cervical cancer is caused due to the Human Papilloma Virus (HPV)

ISO 9001:2008 Certified Journal

Page 3105