Abstract
Ultraviolet radiation (UVR) exposure and family history are major associated risk factors for the development of non-melanoma skin cancer (NMSC). The objective of this study was to develop and validate a multi-parameterized artificial neural network based on available personal health information for early detection of NMSC with high sensitivity and specificity, even in the absence of known UVR exposure and family history. The 1997–2015 NHIS adult survey data used to train and validate our neural network (NN) comprised of 2,056 NMSC and 460,574 non-cancer cases. We extracted 13 parameters for our NN: gender, age, BMI, diabetic status, smoking status, emphysema, asthma, race, Hispanic ethnicity, hypertension, heart diseases, vigorous exercise habits, and history of stroke. This study yielded an area under the ROC curve of 0.81 and 0.81 for training and validation, respectively. Our results (training sensitivity 88.5% and specificity 62.2%, validation sensitivity 86.2% and specificity 62.7%) were comparable to a previous study of basal and squamous cell carcinoma prediction that also included UVR exposure and family history information. These results indicate that our NN is robust enough to make predictions, suggesting that we have identified novel associations and potential predictive parameters of NMSC.
Highlights
Skin cancer is the most commonly occurring malignancy, with basal and squamous cell carcinomas comprising the majority of skin cancer cases[1]
Instead of skin cancer classification based on images, the question we seek to address is: can an artificial neural network (ANN) trained with a large set of health informatics that lacks any ultraviolet radiation (UVR) exposure and family history of non-melanoma skin cancer (NMSC) information be used to predict personal NMSC risk, and if so how well would such a network perform? Our approach is novel and significant as it requires only personal health informatics commonly available in the electronic medical record (EMR) systems, and is a convenient and cost-effective method of evaluating cancer risk for individuals
While not all of these parameters are known to be associated with NMSC, we include them because they are readily available information and the nonlinearity of our ANN means that they can have a bigger impact on our accuracy than traditional statistical methods predict
Summary
Skin cancer is the most commonly occurring malignancy, with basal and squamous cell carcinomas (both of which are classified as non-melanoma) comprising the majority of skin cancer cases[1]. Other studies that rely on Medicare and Medicaid statistics have estimated the number of Americans in 2012 with NMSC was as high as 5.4 million with 3.3 million people being treated that year[2]. Instead of skin cancer classification based on images, the question we seek to address is: can an artificial neural network (ANN) trained with a large set of health informatics that lacks any UVR exposure and family history of NMSC information be used to predict personal NMSC risk, and if so how well would such a network perform? Demographics of the Data Average Age Average BMI Male/Female Ever Smoked Have Emphysema Have Asthma Have Diabetes Mellitus Have Ever Had a Stroke Have Hypertension Average Heart Disease Score White African-American Native American/Alaska Native Asian Multiracial Hispanic Ethnicity Average Number of Times Vigorous Exercise is done at Least Once per Week. Our aim was achieved with an area under the curve (AUC) of 0.8 or above
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
