One way is to set the corresponding coefficients to be 0 depending on which content you are interesting. For example, if you are interesting in microcalcifications in mammography, then you may want to set the coefficients corresponding to low frequency to be 0. 

Greetings. I'm not an expert on wavelets but I recommend you to read the papers from this researcher (Matheus Alvarez). 

https://www.researchgate.net/profile/Matheus_Alvarez

Conference Paper Wavelets in the Computed Evaluation of Mammographic Accredit...

Article Wavelet-based algorithm to the evaluation of contrasted hepa...

Conference Paper Wavelets based algorithm for the evaluation of enhanced liver areas

S. Singh, K. Bovis, An evaluation of contrast enhancement techniques for mammographic breast masses, IEEE Trans. Inf. Technol. Biomed. 9(1)(2005)109–119.

The sub-bands in the wavelet domains can be processed using non-linear transformation functions such that the pixel below a threshold are enhanced and those above are suppressed in an automated (without manual preset/thresholds).

Refer the paper:

V. Bhateja and S. Devi, “Combination of Wavelet Analysis and Non-Linear Enhancement Function for Computer Aided Detection of Microcalcifications,” Proc. of International Conference on Biomedical Engineering and Assistive Technologies (BEATs-2010), Jalandhur (India),  Available: http://www.bmeindia.org/paper/BEATs2010_263.pdf

Try with matlab :After dividing the image into sub bands( if you choose level 1 decomposition you will get 4 sub bands)store those 4 subbands in individual array such as LL,LH,HL,HH.Now you can process the coefficients in each subband.

Example : if your image is of resolution 256 x 256,when you apply level 1 DWT you will get four subbands each of resolution 64x64;

with the help of for loop in matlab you can access those subbands