Lung cancer is a leading cause of death worldwide and smoking is the most preventable risk factor, however some non-smokers also develop lung cancer. There are various techniques used to predict lung cancer occurrence in smokers and non-smokers such as decision tree algorithms like C4.5, CART, and CHAID. Incidence and mortality rates can be calculated using formulas involving the number of new cancer cases or deaths in a population over a given time period. Survival rates are also important for understanding cancer trends and are calculated using observed and expected survival ratios.

1. Running Head: CANCER TREATMENT ALGORITHMS 1
Cancer Treatment Algorithms
Name
Institution

2. Cancer Treatment Algorithms 2
Lung cancer is the most likely cause of death, and its occurrence has increased
worldwide. Lung cancer in men is the most common cancer in men. Various organizations have
suggested that smoking is the most preventable cause of cancer. Cancer can easily spread to
other body parts including the lymph nodes. The incident of a person smoking and has cancer is
90%, and the risk of cancer increases with the amount of tobacco used. There is also a large
population that has died of lung cancer yet they are non-smokers. Therefore there is need to do
data mining for smokers and non-smokers to find the probability of a person smoking having
lung cancer. Data released by several agencies show that up to 16000 Americans die of lung
cancer annually even though they never smoke.
There are many methods used in predicting lung cancer occurrence in both smokers and
non-smokers. There are techniques such as C4.5, CART, AND CHAID. C5, in particular, can be
used in acknowledging noisy data, the algorithm can be used when fitting errors and doing
pruning of data. Using this method, the decisions are easy to make, and the attributes which are
relevant and irrelevant attributes are quickly shown. Most scientist use decision tree algorithms
such as Decision Table, j48, and Naïve Bayes. J48 would the best suit for this work as it uses
independent variables and independent predictors to arrive at the data. The algorithm applies the
ID3 which is also called Iterative Dichotomizer 3 which was developed for WEKA data analysis.
Another method used in estimating the cancer survival is called incidence. Incidence is
calculated as follows:

3. Cancer Treatment Algorithms 3
Incidence = (LCP / TPP) × 10N
LCP = This is the number of the new cancer infections occurred in a given period such as
annually.
TPR= The total number of people at risk.
N = gives the sample population. Therefore mortality can be calculated as follows:
Mortality = (DC / TP) × 10N
DC = This is the number of deaths that occurred in a given period.
TP = Total number of people in the current population
N = 1, 2, 3….
The current period in the algorithm means that the year is considered in the calculation of
the incidence of calculating mortality. Knowledge of cancer survival has enabled the researchers
in estimating the patterns and trends and the fitness of the population. Net survival has shown
that there is little chance of surviving cancer and death from other causes (Cutler et al., 2009).
The survival from cancer does not depend on other causes, in reliable results gives the
approaches used in measuring survival rates. The survival rate is calculated from the real cancer
deaths. Sometimes the cause of death may not be available, and in this case, it may not be
possible to estimate the survival rates of cancer (Duchman, Gao & Miller, 2015). The following
formula can be used in calculating the survival rates:
Relative survival rate = (Observed survival /
Expected survival )×100%
The formulae are very useful in calculating the expected survival ratio for cancer patients.

4. Cancer Treatment Algorithms 4
References
Cutler, S., Ederer, F., Griswold, M., & Greenberg, R. (2009). Survival of Patients With Ovarian
Cancer, Connecticut, 1935–542. JNCI: Journal Of The National Cancer Institute, 24(3),
541-549. http://dx.doi.org/10.1093/jnci/24.3.541
Dutchman, K., Gao, Y., & Miller, B. (2015). Prognostic factors for survival in patients with
Ewing's sarcoma using the surveillance, epidemiology, and results (SEER) program
database. Cancer Epidemiology, 39(2), 189-195.
http://dx.doi.org/10.1016/j.canep.2014.12.012