Early Detection of Oral Cancer Using Smartphone Technology

Study on early detection of Oral cancer using Smartphone -A review from literature Presented by: Meha Saxena Mentor- Dr. Pankaj Talreja IIHMR Delhi 1

Screenshot of Approval Screenshot of Approval 2

Introduction (1/2) Introduction (1/2) Oral cancer is one of the most common cancers worldwide, with high mortality rates. A tumor in the mouth is a sign of oral cancer, which causes pain and suffering. Despite advancements in cancer treatment, the death rate from oral cancer has remained high over the last several decades. For the significant rise in positive survival outcomes, detection of oral cancer at an early stage is very important. 3

Introduction (2/2) Introduction (2/2) A deep machine learning approach has been promoted to improve early diagnosis of cancer. This study performed a systematic review of the published reviews that have examined the smartphone based images using deep machine learning which helps in early detection of cancer. Deep learning algorithms are able to surpass the performance of human experts in many disease recognition scenarios. With the rapid development of both imaging and sensing technologies in camera systems, the ubiquity of smartphones is equipped with higher quality, low-noise, and faster camera modules. 4

Objectives of Objectives of the the Study Study Research Question How has smartphone-based images using deep learning helped in achieving early detection of oral cancer at an early stage which helps patient more likely to be treated successfully? Objective The objective is to find out can smartphone-based images using deep machine learning for detection of oral cancer at an early stage. Importance of detection of oral cancer at an early stage 5

Literature Review Literature Review Review Topic Objective Conclusion Type of Research Aubreville et al. Automatic classification of cancerous tissue in laserendomicroscopy images of the oral cavity using deep learning. Point-of-care, smartphone- based, dual-modality, dual- view, oral cancer screening device with neural network classification for low- resource communities. To detect oral cancer. The deep learning offered automatic detection of oral cancer for effective management of the cancer. Primary Research Uthoff et al. To distinguish between precancerous and cancerous lesions early. Effective management of oral cancer through early detection. Primary Research Chan et al. Texture-map-based branch- collaborative network for oral cancer detection. Deep learning-based survival prediction of oral cancer patients. The detection of oral cancer. The oral cancer was successfully detected. Primary Research Kim et al. Oral cancer survival prediction in patients. Survival prediction can offer a good approach to properly manage oral cancer. Primary Research 6 *Note- Total number of articles reviewed and added in dissertation= 17

Methodology (1/2) Methodology (1/2) Research Design: Descriptive Study Data Type: Secondary Data Data Collection Method: Literature Survey Data Sources: PubMed Search Terms: Cancer, Oral cancer, Early detection, mobile phone, deep learning, machine learning 7

Methodology (2/2) Methodology (2/2) Search strategy- Detailed automated literature searches were performed in PubMed from inception until the end of October 2021. The referencer lists of potentially relevant reviews were searched to ensure that all the potential studies have been included Search Protocol-The search protocol was developed by combining search keywords: [( deep learning AND oral cancer AND smartphone )]. Inclusion criteria- All systematic reviews that considered deep learning for oral cancer detection and smartphone using deep learning were included. This includes reviews that examined the smartphone based images using deep learning on oral cancer detection. Exclusion Criteria- study on deep learning but not directly based on smartphone based images, reviews and editorials were excluded. 8

Methodology Methodology PRISMA flow chart for the included studies 9

Results (1/3) Results (1/3) A total of 20 studies met the eligibility criteria and were included in this review. These studies concluded that smartphone- based imaging helps in early detection of cancer. Smartphone-based images using deep learning for early detection of oral cancer With the rapid development of both imaging and sensing technologies in camera systems, the ubiquity of smartphones is equipped with higher quality, low-noise. The accuracy ranged from 77.89 to 97.51% for images which are captured by mobile phone. This reveals that deep learning models have the potential to assist in the detection of oral cancer * source-https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8786902/ 10

Results (2/3) Results (2/3) Importance of Detection of oral cancer at an early stage Detection of Oral cancer at an early stage is the most effective means to improve survical, reduced morbidity and disfigurement. Early detection of oral cancer has a better prognosis as well as reduced cost of treatment when compared to advanced-stage cancer. 11

Results (3/3) Results (3/3) Figure 1: Five year survival rate for oral cancer by clinical stage 12 *Source: National Cancer Center Japan Cancer Statistics 20

Discussion (1/2) Discussion (1/2) In this systematic review, the utilization of deep machine learning for early detection in oral cancer was examined. The deep learning methodology had been used to analyze various types of medical data for early detection of oral cancer. Figure 2: Image Capturing Method Figure 3: Resampling-Rotation Method 13 *Source: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8397787/

Discussion (2/2) Discussion (2/2) This review showed that the deep machine learning methodology combined with medical imaging data can offer better and helped in early detection of oral cancer. This can significantly assist the clinical management of patients with the disease . Based on the reported accuracy of the deep machine learning techniques in the included studies, it is evident that the deep machine learning technique can play a significant role toward the improved detection of oral cancer and guide clinicians in making informed decisions. Early detection is crucial as it increases the scope for successful treatment. A delay in detection will consecutively cause the cancer cells to spread in the adjacent areas of the affected region, thus increasing the morbidity. More than half of oral cancer patients are diagnosed when presented with advanced lesions. Many are unaware that early diagnosis is key to improved survival, reduced morbidity, disfigurement, duration of hospital stay as well as the over-all cost of treatment. You are not allowed to add slides to this presentation 14

Limitations of the Study Limitations of the Study The main limitation is that the deep learning techniques used different data types in the analyses. So it was difficult to draw conclusions about the effectiveness of these deep learning techniques. Most of the developed deep learning models in the published studies were not externally validated The dataset used to train the model was relatively small in most of the studies. To the best of knowledge, there is a dearth of published studies that have examined the application of machine learning for staging. Therefore, this serves as a potential area of further research in the future. You are not allowed to add slides to this presentation 15

Conclusion Conclusion The deep learning models are poised to detect cancer at eraly stage and predict cancer prognosis more accurately. It is expected that the deep learning techniques can assist in the proper management of oral cancer through improved diagnostic performance, insightful clinical decision making. Thus, the clinicians and patients can spend more time in communication and in making shared decisions to improve the quality of care. You are not allowed to add slides to this presentation 16

References References Chu CS, Lee NP, Ho JW, Choi SW, Thomson PJ. Deep learning for clinical image analyses in oral squamous cell carcinoma: a review. JAMA Otolaryngology Head & Neck Surgery. 2021 Oct 1;147(10):893-900. Zhu W, Xie L, Han J, Guo X. The application of deep learning in cancer prognosis prediction. Cancers. 2020 Mar 5;12(3):603. Uthoff RD, Song B, Sunny S, Patrick S, Suresh A, Kolur T, Keerthi G, Spires O, Anbarani A, Wilder- Smith P, Kuriakose MA. Point-of-care, smartphone-based, dual-modality, dual-view, oral cancer screening device with neural network classification for low-resource communities. PloS one. 2018 Dec 5;13(12):e0207493. Figure1- National Cancer Center Japan Cancer Statistics 20 Figure 2 and 3- https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8397787/ You are not allowed to add slides to this presentation 17

Thank You Any Questions You are not allowed to add slides to this presentation 18

Suggestions to the Organization where the Suggestions to the Organization where the Study was Conducted Study was Conducted Karkinos Healthcare is already working on early detection of oral cancer through mobile app. So , they can use deep machine learning techniques which help them in early detection and prognostication of oral cancer. They can also increase the awareness of Importance of early detection of Oral cancer You are not allowed to add slides to this presentation 19

Dissertation Experiences Dissertation Experiences What did you learn (skill/ topic)? Oral and Cancer and Importance of early detection Product Management Sprint planning Create sprint App design, testing and production Overall self comments on Dissertation Learned about the importance of oral cancer and how deep learning can can help in early detection and prognostication of oral cancer. You are not allowed to add slides to this presentation 20

Pictorial Journey (1/2) Pictorial Journey (1/2) Put 2 of your best photographs here You are not allowed to add slides to this presentation 21

Pictorial Journey (2/2) Pictorial Journey (2/2) Put 2 of your best photographs here You are not allowed to add slides to this presentation 22