Thermal therapies employing nanoparticles generating heat in the presence of alternating electromagnetic field are of high interest to medical community. Currently, therapies employing nanoparticle heating are being investigated on two fronts: (1) they are explored for controlling processes at molecular level; and (2) they are used to produce necrosis in a relatively large biological mass such as a tumor.;Remote heating of nanoparticles for inducing a selective transformation at the molecular/nanoscale level has been proposed relative recently. A series of recent literature reports suggest that nanoparticles are able to produce locally a temperature rise of several degrees above the "bulk" value. However, theoretical predictions based on macroscale heat diffusion theories do not validate these experimental findings. To resolve this apparent disagreement between theory and experimental data, measurement of local temperature rise in the proximity of heated nanoparticles is needed.;This work presents temperature measurements in the vicinity of gold and magnetite nanoparticles suspended in solutions and remotely heated by radio frequency (RF) electromagnetic field. Core-shell semiconductor CdSe quantum dots passivated with a layer of ZnS are used as local temperature sensors. Their temperature is inferred from the shift in the peak of fluorescent emission of the quantum dots. To determine the difference between local and "bulk" temperature, measurements are carried out on two sets of samples: control solutions and conjugated specimens. In the control solutions, the quantum dots are mixed with nanoparticles. Hence, the quantum dots are free to move and they measure an average or "bulk" temperature rise of the sample. In conjugated specimens, the quantum dots are covalently linked to the nanoparticles. Since the distance between covalently bonded quantum dot and nanoparticle is of a few nanometers, they measure the local temperature rise in the proximity of nanoparticle. Both mixed and conjugated specimens have low nanoparticle concentrations, to minimize the heating effect on quantum dots from non-bonded particles in conjugated specimens. From these experiments it is found that measured bulk and local temperature differ by less than 1°C, which is comparable to the experimental error. This suggests that the local heating is negligible, at least for the experimental conditions employed in this work.;While measurements of local temperature rise are important to fundamental understanding of heat transfer in bio-conjugated nano-systems, accurate quantification of "bulk" heating of RF heated nanoparticles is critical for applications such as cancer hyperthermia. The second part of this thesis analyzes the experimental method used to determine the heat generation rate (or specific absorption rate---SAR) of nanoparticles and identifies potential shortcomings associated with the experimental setup and data reduction procedure. While SAR is typically determined from the initial slope of the temperature rise of nanoparticles suspension as function of time, the experimental setups and data reduction techniques vary widely in literature. In many of the SAR installations reported, the coil is directly wounded around the container holding the test solution. Since the magnetic field varies along the radius and the axis of the coil, this configuration may lead to errors associated with non-uniform heat generation. Moreover, since SAR is determined from the transient temperature rise of the sample, it can be affected by the heat capacity of the container. Finally, the data reduction procedure which is often based on polynomial fitting, may also affect the accuracy of reported SAR. All these aspects are illustrated by finite element modeling simulations. In addition to finite element modeling, experiments are also carried out to demonstrate the effect of container heat capacity.