The significant computational overhead associated with the iterative reconstruction process of biomedical optical imaging is one of the major factors that hinder its application in clinical diagnosis, especially for large tissue volumes. Based on previous studies, the paper presents a general purpose biomedical optical image reconstruction model along with components and data flow diagram, and proposes a modified model taking into account the information feedback during the iterative image reconstruction process to control the image resolution in the subdomain of interest, and hence to detect interior targets, such as tumors, from the discrete measurements collected on the surface of clinically relevant tissue. The modified image reconstruction model (MIRM) may provide a means to detect small targets and improve the image quality at relatively low computational cost.