cancerRiskCalc

[molly-pop]

Cancer Risk Calculator: A Novel Application for Tracking Cancer Risk and Preventative Habits

Project Abstract

This document proposes a novel application for a patient-facing web application to assess personal risk factors for cancer development. The user will be able to create an account, and enter data such as age, sex, race, genetic information, and lifestyle factors to determine risk as compared to a typical person of the same age and sex. The application will then provide personalized tips for decreasing risk through preventative measures. It will also provide an interface to track the proposed new habits to lower risk over time.

Conceptual Design

The backend will be programmed using Python. The actual algorithmic calculation of risk will be done in Python referencing the Harvard Cancer Risk report[1] and the relative risk estimates contained within it. Libraries that will be utilized include matplotlib for risk visualization and Sqlite for an embedded backend database to store user data. The frontend framework could be developed using a React framework, or using Python with a framework like Shiny. The preventative information will be sourced from the health.gov APIs.

Proof of Concept

Link to the POC Git Repository: https://github.com/molly-pop/cancerRisk_POC

Background

The web application default will be a text box explaining the nature of the app as a cancer risk calculator and preventative health tracker. It will prompt the user to log in or create an account.

On account creation, they will be prompted to ‘Begin’ the calculator, where they will then click through a form and input data regarding their personal information and habits. At the end of the form, the user will submit, and then be shown a graph containing their risk relative a typical person of the same age and sex, and then be given insight into the preventative measures they can take to minimize their risk going forward. An existing user will be directed to this ending screen showing their risk & preventative measures on login. There will be a separate tab, where the user can input current data on preventative measures they have been implementing, which will load into a chart with a timescale and risk assessment. This will show the user how much they’ve been decreasing their risk over time by incorporating new habits.

This proposed product is similar to the Washington University in St. Louis Siteman Cancer Center calculator [2], and to other disease risk calculators by the NIH, National Cancer Institute, etc. [3], but includes a novel option to track your risk over time with a user account. It will also be more friendly to updates given the use of relevant frameworks and libraries as opposed to the more archaic development seen on many government websites.

Required Resources

No extended hardware beyond a laptop will be necessary for the development of this project. The required software resources have been outlined above in separate sections, but include several programming languages and libraries. The source material for risk assessment is outlined in the above sections but includes data from Harvard, the Siteman Cancer Center, the NIH, and the National Cancer Institute among others.

Project Slide: https://docs.google.com/presentation/d/14An7ddSVnjTQsoMoaLK_l0eM8r6JMT3iIehhI91WQkI/edit?usp=sharing

Bibliography

[1]Siteman Cancer Center, “Your Disease Risk - Prevention - Siteman Cancer Center,” Siteman Cancer Center, 2017. Available: https://siteman.wustl.edu/prevention/ydr/ [2]G. A. Colditz et al., “Harvard report on cancer prevention volume 4: Harvard Cancer Risk Index. Risk Index Working Group, Harvard Center for Cancer Prevention,” Cancer Causes and Control, vol. 11, no. 6, pp. 477–488, 2000, doi: https://doi.org/10.1023/a:1008984432272 [3]National Cancer Institute, “Breast Cancer Risk Assessment Tool,” Breast Cancer Risk Assessment Tool, 2019. Available: https://bcrisktool.cancer.gov/ [4]“APIs for Developers | health.gov,” health.gov. Available: https://health.gov/our-work/national-health-initiatives/health-literacy/consumer-health-content/free-web-content/apis-developers

[kvijay04]

Hello, I think this project proposal has a very useful application in the real world. Cancer is a very deadly disease, and I think that an application that helps prevent is inherently valuable. Additionally, I am very well-versed in coding in python, so I would be able to comfortable fit into this project.

I would personally prefer to work on this project, as I have a lot of experience using python for data visualization, especially matplotlib and seaborn. Additionally, I think that I could greatly contribute to data manipulation and analysis, which could help create a more precise and accurate application. Additionally, because of my experience as a research assistant, I am already knowledgeable of many of the terminology in this field, so I would not need very much training in that regard either. These are the main reasons I think I could greatly contribute to the backend, so I hope you consider me for this project

[ldups]

Hi Molly!

I think that this is a really cool project idea. There is a very clear real-world application, which I think is really cool, and I feel like this project could help a lot of people. I was also impressed with your proof of concept, and I feel like it is already well-developed. It seems like you have a good idea of how to use the APIs, as well as how to develop at least a simple frontend.

I have experience with both Python and frontend JavaScript libraries like React, so I think that I could contribute a lot to this project. I think that it would be really cool to add graphs/charts to allow the user to see their statistics with more context. I have done a lot of plotting in Python with libraries like matplotlib and seaborn, so if you decide on a purely Python application, I think I could contribute a lot on that front. I have also done some plotting in React with the Recharts library. I think it's a really cool library and I would love to continue working with it. I also think that it would be cool to move the frontend to React, and could enable a more responsive application with things like live updating of charts based on user inputs.

Hope we can work on this together!

[ahgoldmeer]

Hi Molly,

I think that this is a great project idea! This is a very prevalent issue and has the potential to create a real impact. Your proof of concept was a really great start for this project, and you clearly know where you want to go with this. I also like how you're using valid medical data to support this idea.

I believe that I could contribute well to this project. I have experience with python that would benefit this project, having worked with data and predictive models previously. For this reason, I think that I would really be able to help with the development of the backend of this application. I additionally have experience with data visualization in python like matplotlib and seaborn, which would integrate well with a Shiny frontend. I think that the ability to save your individual report locally would be nice so that if something comes up, you don't have to rerun it every time to get that information again.

[wSimsT]

Hi Molly,

I really like this project because it is a real life issue. I have cancer in my family tree so it would be very nice to know how likely I am to get it myself. Cancer can often seem unpredictable so this project seems invaluable to address some of those concerns. I also like how your proof of concept looks so far. Your bibliography shows how much real research went into the design of this project. Well done!

I believe that I can contribute to this project as I know python and have experience with APIs through it. I have experience using python for backends and experience with javascript/typescript, which can help with building a frontend. For my job, I do a lot with data processing which means I would be able to help with the backend.

I hope we can work together this semester!

[st-mw]

Hi Molly, I think this is a great idea. I like the fact that every user would get personalized insight into their own personal cancer risk. It has the potential to really help people make positive changes in their lives. The built-in ability to track changes over time is an excellent way to help users stay on track and make progress.

I could contribute to this project because I have experience programming in Python, including processing and visualizing data. I've spent a lot of time working with pandas, numpy, and matplotlib. I think a cool additional feature could be an option for users to sign up for periodic emails with reminders or encouragement.

[magwinst]

I think this project has a lot of potential and offers students the ability to demonstrate their skills towards something directly applicable to people all around the world. Cancer is a devastating disease and tools that can keep patients informed about cancer risk associated with their lifestyle, cheaply, seems like a wonderful idea.

I can definitely help out with the python backend as I am quite skilled in python and am familiar with a plethora of python libraries that are directly applicable to this type of project. I think that I could also potentially take it a step further with my machine learning skills (PyTorch) to enhance the accuracy of the model if you wanted to go down that route.

[NocKtuRn4L-2]

Hi Molly, I really like your idea for a project proposal. Giving people the power to begin preventative care for something as terrible as cancer could help bring a sense of control back into their lives. This could also provide information for them to get to their doctors to have a more informed conversation about their health.

If I were to work on this project I could help with the implementation of the visualization of data and the user interface it is displayed in. I have experience working with python and different machine learning (PyTorch, sci-kit learn) libraries and how to use them to interpret data for risk evaluation.