Inrtroduction
Lung cancer is among the most common variations of the condition. It is extremely dangerous to the person affected by the illness, mainly due to the period in which he or she is most likely to seek medical assistance. Many lung cancers are first detected when they have reached an advanced stage and possibly created metastases. As such, early screening is essential to the prevention and treatment of lung cancer. Databases will likely be beneficial to the effort, as they will enable the comparison of a patient’s information with existing cancer cases. This paper proposes a database that is designed to assist medical workers with the early detection of lung cancer.
Problem Description
As in the case with most variants of the illness, lung cancer has a high comorbidity and mortality rate. Metastatic cancer is almost impossible to cure, and, according to Srivastava (2017), an overwhelming majority of people whose condition proceeds to the stage dies within five years. The author notes that “approximately 70% of patients present with locally advanced or metastatic disease” (p. 187). When the fact that approximately 13% of new cancers are lung cancers, making it the second most common variety (American Cancer Society, 2019), is taken into consideration, the issue becomes more severe. Overall, the lack of early detection measures leads to high patient mortality rates.
Database Impact
The detection of cancer is challenging due to its unknown causes and a lack of a specific agent that can be tested. As such, early detection strategies have to rely on comparisons against existing cases. Chander, Rao, and Rajinikanth (2017) suggest using chest computer tomography images to search for similarities with existing cases that would indicate cancer. D’Cruz, Jadhav, Dighe, Chavan, and Chaudhari (2016) propose an algorithm that can use X-rays, CT images, magnetic resonance imaging results, and other indicators as potential detection sources. The patient’s information can also be used to determine an appropriate treatment through a comparison with the repositories described by Takiguchi (2017). However, all of these options require a database of samples that can be used for comparisons against the patient’s results.
Data Entities
The database will require information about the patients that allows for easy identification and contact. These fields will consist of structured data and include the patient’s first and last name, address, current hospitalization status, phone number, e-mail account, and other information that is necessary for contacting them reliably, such as the names and contact information of their family members. All of this information will be stored in text form and does not require the selection of a specific measuring unit.
The medical information included in the database will vary depending on the detection and treatment approaches used for the screening. As described by Dey, Ashour, Fong, and Bhatt (2019), it will be taken from the EHRs of various divisions and centralized for easier reference. Each test will be stored as structured data in the form of images or measurements accompanied by the date of the test to allow for easier comparisons. The database will likely work with reference sources from outside the U.S., and therefore, it should employ the metric system. The numbers should be recorded using measuring standards that are consistent with the external databases and accessible to computers. For example, size measurements should be recorded in centimeters, millimeters, or other appropriate units as plain numbers, i.e., 180 cm.
Database Organization
The patient will be central to the database, as its purpose is to determine whether he or she has lung cancer. Three separate data sets will be associated with the patient entity: contact information, treatments, and test results. The latter will be organized around a logical variable that shows whether the system suspects that the patient has lung cancer. The field will be informed by several other logical variables that will show whether a specific comparison algorithm has found possible cancer. One positive value in any of these fields will be enough to set the overall result to ‘yes.’ The specific areas will have access to the data they use for reference, though they will not be able to edit it.
Only the overarching patient entity will have access to the contact data, and the information will only come into use when cancer suspicions are confirmed. The specialists who oversee the database can then contact the patient and inform him or her of the potential issue. If the person cannot be reached, the system will allow the users to contact their family members. If a more thorough investigation confirms the presence of cancer, the system may be used to suggest possible treatments. The treatment selection algorithms will have access to the test results they require. The database will update with information from various EHRs regularly, but as its findings are not conclusive, it will not submit data back.
Conclusion
Early screening of lung cancer is a challenging task due to a lack of scientific understanding with regards to the origins and causes of the condition. As such, many patients present with the illness in an advanced stage that is beyond current treatment capacities. However, automated algorithms may detect patterns and help with the task after their reliability is established. A database that draws information from various EHRs, submits it for analysis, enables patient contact for further investigation, and suggests possible treatments may improve early detection rates and patient outcomes for lung cancer.
References
American Cancer Society. (2019). Key statistics for lung cancer. Web.
Chander, M. P., Rao, M. V., & Rajinikanth, T. V. (2017). Detection of lung cancer using digital image processing techniques: A comparative study. International Journal of Medical Imaging, 5(5), 58-62.
D’Cruz, J., Jadhav, A., Dighe, A., Chavan, V., & Chaudhari, J. (2016). Detection of lung cancer using backpropagation neural networks and genetic algorithm. Computer Technology & Applications, 6(5), 823-827.
Dey, N., Ashour, A. S., Fong, S. J., & Bhatt, C. (Eds.). (2019). Healthcare data analytics and management (Vol. 2). San Diego, CA: Academic Press.
Srivastava, S. (Ed.) (2017). Biomarkers in cancer screening and early detection. Oxford, United Kingdom: John Wiley & Sons.
Takiguchi, Y. (Ed.). (2017). Molecular targeted therapy of lung cancer. Singapore, Singapore: Springer.