Objective To compare the errors between data obtained by interval sampling at different time intervals and continuous temperature records, and to analyze the cooling kinetic characteristics of postmortem temperature differences at different measurement organs by monitoring multi-organ synchronous continuous temperatures. Methods Liver, brain, rectal, and ambient temperatures were synchronously and continuously recorded for 72 h in 10 adult male rabbits placed in a climate chamber maintained at 25.4 ℃. A measurement organ was considered to have approached ambient temperature when the absolute difference between the site temperature and the ambient temperature at the corresponding time point within the same batch was ≤0.5 ℃. The time at which each organ first met this criterion was recorded. Rectal temperature, a commonly used measurement in forensic practice, was selected for the interval-sampling analysis. From the 12 h continuous temperature record obtained after each rabbit entered the low temperature-difference stage, rectal and corresponding ambient temperatures were obtained at intervals of 1, 2, 3, and 4 h. Linear interpolation was then used to reconstruct each sampled dataset on the same 1-min time axis as the original record. The mean absolute error （MAE） and root mean square error （RMSE） between each interpolated series and the original continuous rectal temperature-difference record were calculated. Postmortem temperature differences in the liver, brain, and rectum during 0-12 h after death were analyzed according to a first-order kinetic model and analyzed by linear regression to calculate the cooling rate constant K, coefficient of determination R², and half-life of postmortem temperature difference t1/2. Spearman rank correlation analysis was used to examine the relationship between body mass and the K value at each measurement site. Results The first times at which the liver, brain, and rectum approached ambient temperature were （15.41±0.97）h, （18.31±0.76）h, and （21.13±2.21）h, respectively. As the interval between sampled time points increased from 1 to 4 h, the MAE increased from （0.094±0.028）℃ to （0.176±0.045）℃, and the RMSE increased from （0.126±0.034）℃ to (0.230±0.062）℃. During 0-12h after death, the K values of the liver, brain, and rectum were （0.186±0.007）h-1, （0.131±0.011）h-1, and （0.126±0.003）h-1, respectively. The mean R2 values ranged from 0.977 4 to 0.997 0 and the corresponding t1/2 values of the postmortem temperature differences were （3.740±0.154）h, （5.342±0.436）h, and （5.514±0.129）h, respectively. The body mass was negatively correlated with the rectal K value （ρ=-0.726, P<0.05）. Conclusion Multi-organ synchronous continuous temperature monitoring enables continuous recording of the cooling process and subsequent temperature fluctuations of different organs after they approach ambient temperature. The liver, brain, and rectum show different numerical characteristics in the time required to first approach ambient temperature and in the cooling kinetic parameters of postmortem temperature difference. Prolonged sampling intervals increase the errors between the interpolated discrete temperature data and the original continuous records.
