Sorry but I don't deal depthly with NM topics!

it is not the best methods as it is based on many assumptions. 

1. Radiopharmaceutical  distribution through the organs are based on animal studies and these data are just integrated to the human models.

2 For simple calculation, organs are considered as simple geometric object, which is not.

3. Dose is calculated as an average dose over the whole organ.

4. It is assumed that distribution Radiopharmaceutical is uniform through the organs.

5. Depending on type of disease, the activity of organ may change, which are not considered. 

6. The size and shape of body will vary form patient to patient, so great customization is required. 

7. MIRD is also based on old types Radiopharmaceuticals. Distribution,activity and other data of new Radiopharmaceuticals need to be tested with MIRD and should be available for calculation. 

With all those drawbacks, still it is the best method of calculating dose in NM.  Monte Carlo codes can be used to improve calculation and many new software are developed to calculate MIRD more accurately, but still lacks above mentioned facts.

Thanks

Manoj

Manoj covered pretty much all aspects. However, the concept of "target organ" or "critical organ" is no longer the concept you want to pursue, ever since the principle and concept of ALARA (As Low As Reasonably Achievable), taking into account socio-economic conditions was introduced by ICRP (ICRP 26-1977) and revised ALI (Annual Limit on Intake) and DAC (Dose in Air Concentration) values were published for internal intake of radioactive materials in their publication #53. For internal intake of radionuclides, the MIRD formalism, while not being reliable, can be considered a first approximation. However, depending on the organ, deposition and clearance rate constants and radionuclide energy and proximity and the weight for that organ, in some cases the results could support the Critical Organ concept. What you want to do is search for a metabolic kinetic compartmental model or come up with one. Use this model to come up with the rate constants and thus obtain an effective biological life for (1/Teff)=[(1/Tphy)+1/Tbio)] the radionuclide in question. Then it boils down to a simple calculation.

The MIRD system is a good practical system for dosimetry of radionuclides in the human body.  It is well suited to radiation protection applications where organ doses are of interest, and it may not be possible to determine the exact shape and biological performance of the organs of interest.  Idealised organ shapes may be used, and microscopic variations in does are not reckoned.  The weaknesses of the system are in the way real biology may depart from the assumptions.  This is particularly evident in the mass and shape of organs, the uptake and distribution of radionuclides in the organs, the kinetics of uptake and elimination, microdosimetry and so on.  However, it is a good methodology, provided you are alert to its limitations.