I have two annual measurements taken on medical images depicting a lung cancer tumor 's condition. I have likelihood function that taken in the measurement values and estimates malignancy of the tumor. I want to consider cancer or no cancer as the two states of the tumor and the state of the tumor in the first dataset is also known.

Can a particle filter be designed to diagnose the cancer in tumor using the second measurements given the state and measurement in the last/first annual measurements? I also have figured out that an exponential distribution can be used as a state transition model. I have interpolated few data points between the two annual measurements just in case if Particle filter requires some more data for training. But i am unable to put all this information together to make the tumor diagnosis if this is a good case for filtering ?

  • $\begingroup$ Could you please post a part of your data set? $\endgroup$ – Anton May 29 '19 at 7:04
  • $\begingroup$ I have two parameters of the tumor.In the initial/first scan the tumor is non-existent so both parameters are (0,0).After the second annual scan the two values are lets say (0.53,5.566). The first parameter varies between 0.1 and 2.0 and the second parameter varies between 3 and 6. I have developed a likelihood function during my work which takes these two variables and classifies the tumor.Now i want to predict the cancer based on these two scans readings . The initial values are always 0. $\endgroup$ – evolution May 30 '19 at 16:37

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.