BloodNet: An attention-based deep network for accurate, efficient, and costless bloodstain time since deposition inference.

The time since deposition (TSD) of a bloodstain, i.e., the time of a bloodstain formation is an essential piece of biological evidence in crime scene investigation. The practical usage of some existing microscopic methods (e.g., spectroscopy or RNA analysis technology) is limited, as their performance strongly relies on high-end instrumentation and/or rigorous laboratory conditions. This paper presents a practically applicable deep learning-based method (i.e., BloodNet) for efficient, accurate, and costless TSD inference from a macroscopic view, i.e., by using easily accessible bloodstain photos. To this end, we established a benchmark database containing around 50,000 photos of bloodstains with varying TSDs. Capitalizing on such a large-scale database, BloodNet adopted attention mechanisms to learn from relatively high-resolution input images the localized fine-grained feature representations that were highly discriminative between different TSD periods. Also, the visual analysis of the learned deep networks based on the Smooth Grad-CAM tool demonstrated that our BloodNet can stably capture the unique local patterns of bloodstains with specific TSDs, suggesting the efficacy of the utilized attention mechanism in learning fine-grained representations for TSD inference. As a paired study for BloodNet, we further conducted a microscopic analysis using Raman spectroscopic data and a machine learning method based on Bayesian optimization. Although the experimental results show that such a new microscopic-level approach outperformed the state-of-the-art by a large margin, its inference accuracy is significantly lower than BloodNet, which further justifies the efficacy of deep learning techniques in the challenging task of bloodstain TSD inference. Our code is publically accessible via https://github.com/shenxiaochenn/BloodNet. Our datasets and pre-trained models can be freely accessed via https://figshare.com/articles/dataset/21291825.

Application of propofol-remifentanil intravenous general anesthesia combined with regional block in pediatric ophthalmic surgery.

OBJECTIVE: The aim of this study is to observe the anesthetic effect and safety of intravenous anesthesia without muscle relaxant with propofol-remifentanil combined with regional block under laryngeal mask airway in pediatric ophthalmologic surgery. METHODS: A total of 90 undergoing ophthalmic surgery were anesthetized with general anesthesia using the laryngeal mask airway without muscle relaxant. They were randomly divided into two groups: 45 children who received propofol-remifentanil intravenous anesthesia combined with regional block (LG group), and 45 children who received total intravenous anesthesia (G group). The peri-operative circulatory indicators, awakening time after general anesthesia, postoperative analgesic effect and the incidence of anesthesia-related adverse events were respectively compared between the two groups. RESULTS: All the children successfully underwent the surgical procedure. The awakening time after general anesthesia and removal time of laryngeal mask were significantly shorter in the LG group than in the G group (P < 0.05). There was no statistically significant difference in the heart rates in the perioperative period between the two groups (P > 0.05). There was no statistically significant difference in the incidence of intraoperative physical response, respiratory depression, postoperative nausea and vomiting (PONV) and emergence agitation (EA) between the two groups (P > 0.05). The pain score at the postoperative hour 2 was lower in the LG group than in the G group (P < 0.05). CONCLUSION: Propofol-remifentanil intravenous anesthesia combined with long-acting local anesthetic regional block anesthesia, combined with laryngeal mask ventilation technology without muscle relaxants, can be safely used in pediatric eye surgery to achieve rapid and smooth recovery from general anesthesia and better postoperative analgesia. This anesthesia scheme can improve the comfort and safety of children in perioperative period, and has a certain clinical popularization value.

Driver Fatigue Detection Using Heart Rate Variability Features from 2-Minute Electrocardiogram Signals While Accounting for Sex Differences.

Traffic accidents due to fatigue account for a large proportion of road fatalities. Based on simulated driving experiments with drivers recruited from college students, this paper investigates the use of heart rate variability (HRV) features to detect driver fatigue while considering sex differences. Sex-independent and sex-specific differences in HRV features between alert and fatigued states derived from 2 min electrocardiogram (ECG) signals were determined. Then, decision trees were used for driver fatigue detection using the HRV features of either all subjects or those of only males or females. Nineteen, eighteen, and thirteen HRV features were significantly different (Mann-Whitney U test, p < 0.01) between the two mental states for all subjects, males, and females, respectively. The fatigue detection models for all subjects, males, and females achieved classification accuracies of 86.3%, 94.8%, and 92.0%, respectively. In conclusion, sex differences in HRV features between drivers' mental states were found according to both the statistical analysis and classification results. By considering sex differences, precise HRV feature-based driver fatigue detection systems can be developed. Moreover, in contrast to conventional methods using HRV features from 5 min ECG signals, our method uses HRV features from 2 min ECG signals, thus enabling more rapid driver fatigue detection.

Dissecting the microbial community structure of internal organs during the early postmortem period in a murine corpse model.

BACKGROUND: Microorganisms distribute and proliferate both inside and outside the body, which are the main mediators of decomposition after death. However, limited information is available on the postmortem microbiota changes of extraintestinal body sites in the early decomposition stage of mammalian corpses. RESULTS: This study investigated microbial composition variations among different organs and the relationship between microbial communities and time since death over 1 day of decomposition in male C57BL/6 J mice by 16S rRNA sequencing. During 1 day of decomposition, Agrobacterium, Prevotella, Bacillus, and Turicibacter were regarded as time-relevant genera in internal organs at different timepoints. Pathways associated with lipid, amino acid, carbohydrate and terpenoid and polyketide metabolism were significantly enriched at 8 h than that at 0.5 or 4 h. The microbiome compositions and postmortem metabolic pathways differed by time since death, and more importantly, these alterations were organ specific. CONCLUSION: The dominant microbes differed by organ, while they tended toward similarity as decomposition progressed. The observed thanatomicrobiome variation by body site provides new knowledge into decomposition ecology and forensic microbiology. Additionally, the microbes detected at 0.5 h in internal organs may inform a new direction for organ transplantation.

A Lightweight Sentiment Analysis Framework for a Micro-Intelligent Terminal.

Sentiment analysis aims to mine polarity features in the text, which can empower intelligent terminals to recognize opinions and further enhance interaction capabilities with customers. Considerable progress has been made using recurrent neural networks or pre-trained models to learn semantic representations. However, recently published models with complex structures require increasing computational resources to reach state-of-the-art (SOTA) performance. It is still a significant challenge to deploy these models to run on micro-intelligent terminals with limited computing power and memory. This paper proposes a lightweight and efficient framework based on hybrid multi-grained embedding on sentiment analysis (MC-GGRU). The gated recurrent unit model is designed to incorporate a global attention structure that allows contextual representations to be learned from unstructured text using word tokens. In addition, a multi-grained feature layer can further enrich sentence representation features with implicit semantics from characters. Through hybrid multi-grained representation, MC-GGRU achieves high inference performance with a shallow structure. The experimental results of five public datasets show that our method achieves SOTA for sentiment classification with a trade-off between accuracy and speed.

Analysis of Postmortem Intestinal Microbiota Successional Patterns with Application in Postmortem Interval Estimation.

Microorganisms play a vital role in the decomposition of vertebrate remains in natural nutrient cycling, and the postmortem microbial succession patterns during decomposition remain unclear. The present study used hierarchical clustering based on Manhattan distances to analyze the similarities and differences among postmortem intestinal microbial succession patterns based on microbial 16S rDNA sequences in a mouse decomposition model. Based on the similarity, seven different classes of succession patterns were obtained. Generally, the normal intestinal flora in the cecum was gradually decreased with changes in the living conditions after death, while some facultative anaerobes and obligate anaerobes grew and multiplied upon oxygen consumption. Furthermore, a random forest regression model was developed to predict the postmortem interval based on the microbial succession trend dataset. The model demonstrated a mean absolute error of 20.01 h and a squared correlation coefficient of 0.95 during 15-day decomposition. Lactobacillus, Dubosiella, Enterococcus, and the Lachnospiraceae NK4A136 group were considered significant biomarkers for this model according to the ranked list. The present study explored microbial succession patterns in terms of relative abundances and variety, aiding in the prediction of postmortem intervals and offering some information on microbial behaviors in decomposition ecology.

The Expression of MDM2 Gene Promoted Chondrocyte Proliferation in Rats with Osteoarthritis via the Wnt/ß-Catenin Pathway.

This study aimed to investigate the regulatory mechanism of MDM2 gene expression on cartilage cell proliferation in Osteoarthritis (OA) rats. For this purpose, 22 SD rats were randomly divided into normal control (10 cases) and treated (12 cases) groups. Treated group was used for OA modelling with the modified Hulth method. After a week, RT-PCR was used to detect MDM2 in cartilage tissue of rats, Wnt 1, Wnt 3 a, Wnt 10 b and ß-catenin genes mRNA expression. Rat chondrocytes were isolated and cultured, and the recombinant eukaryotic expression vector pcDNA3.1 myc-siRNA-MDM2-ß-catenin and co-expression plasmid pcDNA3.1 myc-siRNA-MDM2-ß-catenin was used to transfect chondrocytes and the proliferation and related gene expression levels of the transfected chondrocytes were detected by MTT method and RT-PCR. The results showed that compared with the control group, MDM2, Wnt 1, Wnt 3 a, Wnt 10b and ß-catenin genes in OA rat cartilage constructed by Hulth method were increased (p<0.05). The pcDNA3.1 myc-beta-catenin transfection slowed down the proliferation of OA chondrocytes, different from the non-transfected OA group (p<0.001), and increased Wnt 1, Wnt 3a, Wnt 10b and ß-catenin genes expression compared with the Control group (p<0.05), but did not affect the expression of MDM2. The transfection of siRNA-MDM2 was opposite to pcDNA3.1 myc-ß-catenin. The co-expression plasmid pcDNA3.1 myc-siRNA-MDM2-beta-catenin transfection did not affect the proliferation of OA chondrocytes. In general, the high expression of MDM2 in OA rats restricts the proliferation of chondrocytes, which may be related to the main pathogenesis of the occurrence and development of OA in vivo, and the regulation of MDM2 on the proliferation of chondrocytes may be achieved through the Wnt/ ß-catenin pathway.

Research on a Real-Time Estimation Method of Vehicle Sideslip Angle Based on EKF.

In this article, a real-time vehicle sideslip angle state observer is proposed, which is based on the EKF algorithm. Firstly, based on a 2-DOF dynamical model and the tire lateral force model, the vehicle sideslip angle state observer model with a self-adapted truncation procedure is established by combining the EKF and the least squares methods. The calculation of the Jacobi matrix in the time domain is transformed into a calculation in the frequency domain. A self-adapted update noise estimation method and an initial value setting strategy are proposed as well. Finally, a hardware-in-the-loop simulation is carried out by Matlab/Simulink-CarSim-dSPACE, and the real-time reliability of the estimation method is verified and analyzed by RMSE.

An Emerging Strategy for Muscle Evanescent Trauma Discrimination by Spectroscopy and Chemometrics.

Trauma is one of the most common conditions in the biomedical field. It is important to identify it quickly and accurately. However, when evanescent trauma occurs, it presents a great challenge to professionals. There are few reports on the establishment of a rapid and accurate trauma identification and prediction model. In this study, Fourier transform infrared spectroscopy (FTIR) and microscopic spectroscopy (micro-IR) combined with chemometrics were used to establish prediction models for the rapid identification of muscle trauma in humans and rats. The results of the average spectrum, principal component analysis (PCA) and loading maps showed that the differences between the rat muscle trauma group and the rat control group were mainly related to biological macromolecules, such as proteins, nucleic acids and carbohydrates. The differences between the human muscle trauma group and the human control group were mainly related to proteins, polysaccharides, phospholipids and phosphates. Then, a partial least squares discriminant analysis (PLS-DA) was used to evaluate the classification ability of the training and test datasets. The classification accuracies were 99.10% and 93.69%, respectively. Moreover, a trauma classification and recognition model of human muscle tissue was constructed, and a good classification effect was obtained. The classification accuracies were 99.52% and 91.95%. In conclusion, spectroscopy and stoichiometry have the advantages of being rapid, accurate and objective and of having high resolution and a strong recognition ability, and they are emerging strategies for the identification of evanescent trauma. In addition, the combination of spectroscopy and stoichiometry has great potential in the application of medicine and criminal law under practical conditions.

Species identification of semen stains by ATR-FTIR spectroscopy.

Semen stains are the most important biological evidence when identifying the aggressor in sexual assault cases. Current detecting assays of semen stains species identification were not confirmative enough. In this study, we investigated the potential of species identification of semen stains by using attenuated total reflection (ATR) Fourier transform infrared (FTIR) spectroscopy combined with advanced chemometrics methods. The effect of substrates types and time since deposition (TSD) were considered in the study. A partial least squares-discriminant analysis (PLS-DA) classification model was established which demonstrated complete separation between human and other species (rabbit, dog, boar, bull, and ram). Validation was conducted which showed prediction abilities with 100% accuracy. Additionally, we found species identification could be achieved without sperm cells which proved ability of spectroscopic methods detecting the semen samples from the case of azoospermia. This work provides a powerful and practical tool for species identification of semen stains in real forensic casework.

Identification of antemortem and postmortem fractures in a complex environment by FTIR spectroscopy based on a rabbit tibial fracture self-control model.

The identification of antemortem and postmortem fractures is a critical and challenging task for forensic researchers. Based on our preliminary studies, we explored whether the combination of Fourier transform infrared spectroscopy (FTIR) and chemometrics can identify antemortem and postmortem fractures in complex environments. The impacts of the four environments on the bone spectrum were analyzed by principal component analysis (PCA). It was found that the bone degradation rate in the submerged and ground surface (GS) environments was higher than that in the buried and constant temperature and moisture (CTM) environments. Additionally, the bone degradation rate in buried environment higher than that in the CTM environment. The average spectrum, PCA and partial least squares discriminant analysis (PLS-DA) results all revealed that there were significant differences between the antemortem fracture and the remaining three groups in a complex environment. Compared with the antemortem fracture, the antemortem fracture control (AFC) and postmortem fracture control (PFC) tended to be more similar to the postmortem fracture. According to the loading plot, amide I and amide II were the main components that contributed to the identification of the antemortem fracture, AFC, postmortem fracture, and PFC. Finally, we established a differential model for the antemortem and postmortem fractures (an accuracy of 96.9%), and a differentiation model for the antemortem fracture, AFC, postmortem fracture, and PFC (an accuracy of 87.5%). In conclusion, FTIR spectroscopy is a reliable tool for the identification of antemortem and postmortem fractures in complex environments.

Encapsulated Polyethyleneimine Enables Synchronous Nanostructure Construction and In Situ Functionalization of Nanofiltration Membranes.

Highly permselective nanostructured membranes are desirable for the energy-efficient molecular sieving on the subnanometer scale. The nanostructure construction and charge functionalization of the membranes are generally carried out step by step through the conventional layer-by-layer coating strategy, which inevitably brings about a demanding contradiction between the permselective performance and process efficiency. For the first time, we report the concurrent construction of the well-defined molecular sieving architectures and tunable surface charges of nanofiltration membranes through precisely controlled release of the nanocapsule decorated polyethyleneimine and carbon dioxide. This novel strategy not only substantially shortens the fabrication process but also leads to impressive performance (permeance up to 37.4 L m-2 h-1 bar-1 together with a rejection 98.7% for Janus Green B-511 Da) that outperforms most state-of-art nanofiltration membranes. This study unlocks new avenues to engineer next-generation molecular sieving materials simply, precisely, and cost efficiently.

Predicting postmortem interval based on microbial community sequences and machine learning algorithms.

Microbes play an essential role in the decomposition process but were poorly understood in their succession and behaviour. Previous researches have shown that microbes show predictable behaviour that starts at death and changes during the decomposition process. Research of such behaviour enhances the understanding of decomposition and benefits estimating the postmortem interval (PMI) in forensic investigations, which is critical but faces multiple challenges. In this study, we combined microbial community characterization, microbiome sequencing from different organs (i.e. brain, heart and cecum) and machine learning algorithms [random forest (RF), support vector machine (SVM) and artificial neural network (ANN)] to investigate microbial succession pattern during corpse decomposition and estimate PMI in a mouse corpse system. Microbial communities exhibited significant differences between the death point and advanced decay stages. Enterococcus faecalis, Anaerosalibacter bizertensis, Lactobacillus reuteri, and so forth were identified as the most informative species in the decomposition process. Furthermore, the ANN model combined with the postmortem microbial data set from the cecum, which was the best combination among all candidates, yielded a mean absolute error of 1.5 ± 0.8 h within 24-h decomposition and 14.5 ± 4.4 h within 15-day decomposition. This integrated model can serve as a reliable and accurate technology in PMI estimation.

Molecular characterization of gut microbial shift in SD rats after death for 30 days.

To observe the temporal shifts of the intestinal microbial community structure and diversity in rats for 30 days after death. Rectal swabs were collected from rats before death (BD) and on day 1, 5, 10, 15, 20, 25, and 30 after death (AD). Bacteria genomic DNA was extracted and V3 + V4 regions of 16S rRNA gene were amplified by PCR. The amplicons were sequenced at Illumina MiSeq sequencing platform. The bacterial diversity and richness showed similar results from day 1 to 5 and day 10 to 25 all presenting downtrend, while from day 5 to 10 showed slightly increased. The relative abundance of Firmicutes and Proteobacteria displayed inverse variation in day 1, 5, 10 and that was the former decreased, the latter increased. Bacteroidetes, Spirochaete and TM7 in day 15, 20, 25, 30 was significantly decline comparing with BD. Enterococcus and Proteus displayed reduced trend over day 1, 5, 10 and day 10, 15, 20, 25, respectively, while Sporosarcina showed obvious elevation during day 15, 20, 25. Accordingly, there was a certain correlation between intestinal flora succession and the time of death. The results suggested that intestinal flora may be potential indicator to aid estimation of post-mortem interval (PMI).

Evaluating the effects of causes of death on postmortem interval estimation by ATR-FTIR spectroscopy.

Estimating postmortem interval (PMI) is one of the most challenging tasks in forensic practice due to the effects of many factors. Here, attenuated total reflection Fourier transform infrared (ATR-FTIR) spectroscopy combined with chemometrics was utilized to evaluate the effects of causes of death when estimating PMI and to establish a partial least square (PLS) regression model, which can precisely predict PMI under different causes of death. First, the sensitivities to causes of death (brainstem injury, mechanical asphyxia, and hemorrhage shock) of seven kinds of organs were evaluated based on their degrees of cohesion and separation. Then, the liver was selected as the most sensitive organ to establish a PMI estimation model to compare the predicted deviations from different causes of death. It turns out that the cause of death has no significant effect on estimating PMI. Next, a PLS regression model was built with kidney tissues, which have the lowest sensitivity, and this model showed a satisfactory predictive ability and wide applicability. This study demonstrates the feasibility of using ATR-FTIR spectroscopy in conjunction with chemometrics as a powerful alternative for detecting changes in biochemistry and estimating PMI. A new perspective was also provided for evaluating the effect of causes of death when predicting PMI.

How Geographical Isolation and Aging in Place Can Be Accommodated Through Connected Health Stakeholder Management: Qualitative Study With Focus Groups.

BACKGROUND: In remote areas, connected health (CH) is needed, but as local resources are often scarce and the purchasing power of residents is usually poor, it is a challenge to apply CH in these settings. In this study, CH is defended as a technological solution for reshaping the direction of health care to be more proactive, preventive, and precisely targeted-and thus, more effective. OBJECTIVE: The objective of this study was to explore the identity of CH stakeholders in remote areas of Taiwan and their interests and power in order to determine ideal strategies for applying CH. We aimed to explore the respective unknowns and discover insights for those facing similar issues. METHODS: Qualitative research was conducted to investigate and interpret the phenomena of the aging population in a remote setting. An exploratory approach was employed involving semistructured interviews with 22 participants from 8 remote allied case studies. The interviews explored perspectives on stakeholder arrangements, including the power and interests of stakeholders and the needs of all the parties in the ecosystem. RESULTS: Results were obtained from in-depth interviews and focus groups that included identifying the stakeholders of remote health and determining how they influence its practice, as well as how associated agreements bring competitive advantages. Stakeholders included people in government sectors, industrial players, academic researchers, end users, and their associates who described their perspectives on their power and interests in remote health service delivery. Specific facilitators of and barriers to effective delivery were identified. A number of themes, such as government interests and power of decision making, were corroborated across rural and remote services. These themes were broadly grouped into the disclosure of conflicts of interest, asymmetry in decision making, and data development for risk assessment. CONCLUSIONS: This study contributes to current knowledge by exploring the features of CH in remote areas and investigating its implementation from the perspectives of stakeholder management. It offers insights into managing remote health through a CH platform, which can be used for preliminary quantitative research. Consequently, these findings could help to more effectively facilitate diverse stakeholder engagement for health information sharing and social interaction.

A Wireless High-Sensitivity Fetal Heart Sound Monitoring System.

In certain cases, the condition of the fetus can be revealed by the fetal heart sound. However, when the sound is detected, it is mixed with noise from the external environment as well as internal disturbances. Our exclusive sensor, which was constructed of copper with an enclosed cavity, was designed to prevent external noise. In the sensor, a polyvinylidene fluoride (PVDF) piezoelectric film, with a frequency range covering that of the fetal heart sound, was adopted to convert the sound into an electrical signal. The adaptive support vector regression (SVR) algorithm was proposed to reduce internal disturbance. The weighted-index average algorithm with deviation correction was proposed to calculate the fetal heart rate. The fetal heart sound data were weighted automatically in the window and the weight was modified with an exponent between windows. The experiments show that the adaptive SVR algorithm was superior to empirical mode decomposition (EMD), the self-adaptive least square method (LSM), and wavelet transform. The weighted-index average algorithm weakens fetal heart rate jumps and the results are consistent with reality.

Human and non-human bone identification using FTIR spectroscopy.

Human and non-human identification of unknown skeletal remains is of great importance in forensic and anthropologic contexts. However, the traditional morphological methods for bone species identification are subjective or time-consuming. Here, we utilized Fourier transform infrared (FTIR) spectroscopy and chemometric methods to determinate the spectral variances between human and non-human (i.e., pig, goat, and cow) bones. To simulate real forensic situations as much as possible, fresh, boiled, and decomposed bones were included in this study. Principal component analysis (PCA) results illustrated pig bones were more sensitive to the environmental and external factors than other species studied in this work. Thus, pig bone might not be a suitable proxy for human bone in the study of postmortem changes. More importantly, score plots of PCA results showed clear separation with a slight overlap between the human and non-human fresh bones, but it failed to distinguish the boiled and decomposed bones. Then, partial least squares discriminant analysis (PLS-DA) was employed, and both internal and external validations were conducted to assess its classification ability, which resulted in 99.72 and 99.53% accuracy, respectively. According to the loading plots of PCA and PLS-DA, the spectral diversity was mainly due to the inorganic portion (i.e., carbonates and phosphates), which can remain relatively stable under various conditions. As such, our results illustrate that FTIR spectroscopy could serve as a reliable tool to assist in bone species determination and also has great potential in real forensic cases with natural conditions.

An investigation on annular cartilage samples for post-mortem interval estimation using Fourier transform infrared spectroscopy.

Many attempts have been made to estimate the post-mortem interval (PMI) using bioanalytical methods based on multiple biological samples. Cartilage tissues could be used as an alternative for this purpose because their rate of degradation is slower than that of other soft tissue or biofluid samples. In this study, we applied Fourier transform infrared (FTIR) spectroscopy to acquire bioinformation from human annular cartilages within 30 days post-mortem. Principal component analysis (PCA) showed that sex and causes of death have almost no impact on the overall spectral variations caused by post-mortem changes. With pre-processing approaches, several predicted models were established using a conventional machine learning method, known as the partial least square (PLS) regression. The best model achieved a satisfactory prediction with a low error of 1.49 days using the second derivative transform of 3-point smoothing and extended multiplicative scatter correction (EMSC), and the spectral regions from proteins and carbohydrates contributed greatly to the PMI prediction. This study demonstrates the feasibility of cartilage-based FTIR analysis for PMI estimation. Further work will introduce advanced algorithms for more accurate and precise PMI prediction.

Postmortem diagnosis of fatal hypothermia/hyperthermia by spectrochemical analysis of plasma.

Postmortem diagnosis of extreme-weather-related deaths is a challenging forensic task. Here, we present a state-of-the-art study that employed attenuated total reflection (ATR) Fourier transform infrared (FTIR) spectroscopy in combination with Chemometrics for postmortem diagnosis of fatal hypothermia/hyperthermia by biochemical investigation of plasma in rats. The results of principal component analysis (PCA) and spectral analysis revealed that plasma samples from the fatal hypothermia, fatal hyperthermia, and control groups, are substantially different from each other based on the spectral variations associated with the lipid, carbohydrate and nucleic acid components. Two partial least squares-discriminant analysis (PLS-DA) classification models (hypothermia-nonhypothermia and hyperthermia-nonhyperthermia binary models) with a 100% accuracy rate were constructed. Subsequently, internal cross-validation was performed to assess the robustness of these two models, which resulted in 98.1 and 100% accuracy. Ultimately, classification predictions of 42 unknown plasma samples were performed by these two models, and both models achieved 100% accuracy. Additionally, our results demonstrated that hemolysis and postmortem hypothermic/hyperthermic effects did not weaken the prediction ability of these two classification models. In summary, this work demonstrates ATR-FTIR spectroscopy's great potential for postmortem diagnosis of fatal hypothermia/hyperthermia.