Brain fingerprinting and cognitive behavior predicting using functional connectome of high inter-subject variability.

The functional connectivity (FC) graph of the brain has been widely recognized as a ``fingerprint'' that can be used to identify individuals from a group of subjects. Research has indicated that individual identification accuracy can be improved by eliminating the impact of shared information among individuals. However, current research extracts not only shared information of inter-subject but also individual-specific information from FC graphs, resulting in incomplete separation of shared information and fingerprint information among individuals, leading to lower individual identification accuracy across all functional magnetic resonance imaging (fMRI) states session pairs and poor cognitive behavior prediction performance. In this paper, we propose a method to enhance inter-subject variability combining conditional variational autoencoder (CVAE) network and sparse dictionary learning (SDL) module. By embedding fMRI state information in the encoding and decoding processes, the CVAE network can better capture and represent the common features among individuals and enhance inter-subject variability by residual. Our experimental results on Human Connectome Project (HCP) data show that the refined connectomes obtained by using CVAE with SDL can accurately distinguish an individual from the remaining participants. The success accuracies reached 99.7 % and 99.6 % in the session pair rest1-rest2 and reverse rest2-rest1, respectively. In the identification experiment involving task-task combinations carried out on the same day, the identification accuracies ranged from 94.2 % to 98.8 %. Furthermore, we showed the Frontoparietal and Default networks make the most significant contributions to individual identification and the edges that significantly contribute to individual identification are found within and between the Frontoparietal and Default networks. Additionally, high-level cognitive behaviors can also be better predicted with the obtained refined connectomes, suggesting that higher fingerprinting can be useful for resulting in higher behavioral associations. In summary, our proposed framework provides a promising approach to use functional connectivity networks for studying cognition and behavior, promoting a deeper understanding of brain functions.

Discovering individual fingerprints in resting-state functional connectivity using deep neural networks.

Non-negligible idiosyncrasy due to interindividual differences is an ongoing issue in resting-state functional MRI (rfMRI) analysis. We show that a deep neural network (DNN) can be employed for individual identification by learning important features from the time-varying functional connectivity (FC) of rfMRI in the Human Connectome Project. We employed the trained DNN to identify individuals from an independent dataset acquired at our institution. The results revealed that the DNN could successfully identify 300 individuals with an error rate of 2.9% using 15 s time-window and 870 individuals with an error rate of 6.7%. A trained DNN with nonlinear hidden layers led to the proposal of the "fingerprint of FC" (fpFC) as representative edges of individual FC. The fpFCs for individuals exhibited commonly important and individual-specific edges across time-window lengths (from 5 min to 15 s). Furthermore, the utility of our model for another group of subjects was validated, supporting the feasibility of our technique in the context of transfer learning. In conclusion, our study offers an insight into the discovery of the intrinsic mode of the human brain using whole-brain resting-state FC and DNNs.

Population genetic analyses of Eastern Chinese Han nationality using ForenSeq™ DNA Signature Prep Kit.

Currently, the most commonly used method for human identification and kinship analysis in forensic genetics is the detection of length polymorphism in short tandem repeats (STRs) using polymerase chain reaction (PCR) and capillary electrophoresis (CE). However, numerous studies have shown that considerable sequence variations exist in the repeat and flanking regions of the STR loci, which cannot be identified by CE detection. Comparatively, massively parallel sequencing (MPS) technology can capture these sequence differences, thereby enhancing the identification capability of certain STRs. In this study, we used the ForenSeq™ DNA Signature Prep Kit to sequence 58 STRs and 94 individual identification SNPs (iiSNPs) in a sample of 220 unrelated individuals from the Eastern Chinese Han population. Our aim is to obtain MPS-based STR and SNP data, providing further evidence for the study of population genetics and forensic applications. The results showed that the MPS method, utilizing sequence information, identified a total of 486 alleles on autosomal STRs (A-STRs), 97 alleles on X-chromosome STRs (X-STRs), and 218 alleles on Y-chromosome STRs (Y-STRs). Compared with length polymorphism, we observed an increase of 260 alleles (157, 31, and 72 alleles on A-STRs, X-STRs, and Y-STRs, respectively) across 36 STRs. The most substantial increments were observed in DYF387S1 and DYS389II, with increases of 287.5% and 250%, respectively. The most increment in the number of alleles was found at DYF387S1 and DYS389II (287.5% and 250%, respectively). The length-based (LB) and sequence-based (SB) combined random match probability (RMP) of 27 A-STRs were 6.05E-31 and 1.53E-34, respectively. Furthermore, other forensic parameters such as total discrimination power (TDP), cumulative probability of exclusion of trios (CPEtrio), and duos (CPEduo) were significantly improved when using the SB data, and informative data were obtained for the 94 iiSNPs. Collectively, these findings highlight the advantages of MPS technology in forensic genetics, and the Eastern Chinese Han genetic data generated in this study could be used as a valuable reference for future research in this field.

Screening a new set of microhaplotypes in exonic regions for sample identity testing and paternity testing during whole exome sequencing analysis.

Whole exome sequencing (WES) is widely used in clinical diagnosis. Before obtaining an accurate diagnosis, it is essential to conduct sample identity testing and paternity testing on trio samples. Currently, there is a lack of optimal genetic markers for these purposes, with limited literature available in this area. Microhaplotypes (MHs) are promising genetic markers due to their high polymorphism, low mutation rate, short amplified fragments, absence of stutter and amplification bias. These characteristics make them suitable for sample tracking and paternity testing during WES analysis. In this study, we screened out a set of polymorphic MHs in exonic regions for the above purposes. The results showed that the power of discrimination (PD) and probability of exclusion (PE) of this set of markers ranged from 0.2682 to 0.8878 and 0.0178 to 0.4583, respectively. Both the cumulative power of discrimination (CPD) and cumulative probability of exclusion (CPE) exceeded 0.999999, indicating the great value of these markers in paternity testing and individual identification in the study population. However, these markers had the effective number of alleles (Ae) values ranging from 1.1784 to 3.8727 (average 2.1805) and informativeness (In) values ranging from 0.0151 to 0.2209 (average 0.0766), showing limited value in DNA mixture analysis and biogeographical ancestry inference.

Genetic polymorphism analysis and forensic application evaluation of 57 insertion/deletion polymorphisms from Yi ethnic group in Yunnan.

BACKGROUND: As a new kind of diallelic genetic marker, insertion/deletion (InDel) polymorphisms have recently been used in forensic science. However, there are relatively few studies on the forensic evaluation of InDel genetic polymorphisms from different populations. AIM: The aim of the present work is to assess the genetic polymorphism and forensic applicability of 57 InDels from the Yi ethnic group and explore the genetic background of this group. SUBJECTS AND METHODS: A total sample of 122 unrelated individuals of Yi group from the Yunnan province were genotyped by the AGCU indel 60 Kit. Multiplex population genetic analyses on the same 57 InDels were carried out among the Yunnan Yi group and 29 reference populations. RESULTS: The average allele frequency of these loci in the Yi ethnic group was 0.485. Heterozygosity, polymorphism information content, and the power of discrimination were 0.477, 0.362, and 0.612, respectively. The combined power of discrimination and the combined power of exclusion reached to 0.99999999999999999669 and 0.999962965, respectively. The results showed that 57 InDels polymorphisms have high genetic polymorphisms in the Yi ethnic group. CONCLUSIONS: The 57 InDels could be used for forensic individual identification, paternity testing, and intercontinental population discrimination, with the potential for use in biogeographic ancestry inference.

Establishment and Application of a 42-plex Microhaplotype Assay in Forensic Genetics.

OBJECTIVES: To establish and forensically verify a 42 microhaplotypes (mircohaps, MHs) multiplex assay system based on next-generation sequencing (NGS), and to explore the application value of this system in the practice of forensic genetics. METHODS: A total of 42 highly polymorphic MHs were selected from previous studies, and sequenced by the MiSeq FGxTM platform to verify the repeata-bility, sensitivity, specificity, stability, and mixture analysis ability of the detection system. Through population genetic investigation of 102 unrelated Chinese Han individuals in Liyang City, Jiangsu Province, China, the application value of this system in forensic genetics was evaluated. RESULTS: The sequencing repeatability of the 42-plex MHs assay was 100% and the sensitivity was as low as 0.062 5 ng. The system had the ability to withstand the interference of indigo (≤2 500 ng/µL), humic acid (≤9 ng/µL), hemoglobin(≤20 µmol), and urea (≤200 ng/µL) and to detect mixtures of 2 people (1∶19), 3 people (1∶1∶9) and 4 people (1∶1∶1∶9). Based on 102 individual data, the combined power of discrimination and the combined power of exclusion were 1-3.45×10-30 and 1-3.77×10-11, respectively, and the average effect value of alleles was 2.899. CONCLUSIONS: The 42-plex MHs assay was successfully established in this study and this system has high repeatability and sensitivity, good anti-jamming ability and mixture analysis ability. The 42 MHs are highly polymorphism and have good application value in individual identification and paternity testing.

Research Progress on Biological Evidence Identification in Fire Scenes.

Biological evidence is relatively common evidence in criminal cases, and it has strong probative power because it carries DNA information for individual identification. At the scene of fire-related cases, the complex thermal environment, the escape of trapped people, the firefighting and rescue operations, and the deliberate destruction of criminal suspects will all affect the biological evidence in the fire scene. Scholars at home and abroad have explored and studied the effectiveness of biological evidence identification in fire scenes, and found that the blood stains, semen stains, bones, etc. are the main biological evidence which can be easily recovered with DNA in fire scenes. In order to analyze the research status and development trend of biological evidence in fire scenes, this paper systematically sorts out the relevant research, mainly including the soot removal technology, appearance method of typical biological evidence, and possibility of identifying other biological evidence. This paper also prospects the next step of research direction, in order to provide reference for the identification of biological evidence and improve the value of biological evidence in fire scenes.

The Latest Research Progress on Cell-Free DNA and Prospects of Its Forensic Application.

In recent years, with the continuous progress of DNA extraction and detection technology, cell-free DNA(cfDNA)has been widely used in the life science field, and its potential application value in forensic identification is becoming more and more obvious. This paper reviews the concept, formation mechanism, and classification of cfDNA, etc., and describes the latest research progress of cfDNA in personal identification of crime scene touch DNA samples and non-invasive prenatal paternity testing (NIPPT). Meanwhile, this paper summarizes the potential application of cfDNA in injury inference, and discusses the advantages and disadvantages of common cfDNA analysis methods and techniques, and its application prospects, to provide a new idea for the wide application of cfDNA in the field of forensic science.

Development of Genetic Markers for Sex and Individual Identification of the Japanese Giant Flying Squirrel (Petaurista leucogenys) by an Efficient Method Using High-Throughput DNA Sequencing.

DNA markers that detect differences in the number of microsatellite repeats can be highly effective for genotyping individuals that lack differences in external morphology. However, isolation of sequences with different microsatellite repeat numbers between individuals has been a time-consuming process in the development of DNA markers. Individual identification of Japanese giant flying squirrels (Petaurista leucogenys) has been challenging because this species is arboreal and nocturnal and exhibits little to no morphological variation between individuals. In this study, we developed DNA markers for sex and individual identification of this species by an efficient method using high-throughput DNA sequence data. Paired-end 5 Gb (2 × 250 bp) and 15 Gb (2 × 150 bp) genome sequences were determined from a female and a male Japanese giant flying squirrel, respectively. We searched SRY and XIST genes located on Y and X chromosomes, respectively, from high-throughput sequence data and designed primers to amplify these genes. Using these primer sets, we succeeded to identify the sex of individuals. In addition, we selected 12 loci containing microsatellites with different numbers of repeats between two individuals from the same data set, and designed primers to amplify these sequences. Twenty individuals from nine different locations were discriminated using these primer sets. Furthermore, both sex and microsatellite markers were amplified from DNA extracted non-invasively from single fecal pellet samples. Based on our results for flying squirrels, we expect our efficient method for developing non-invasive high-resolution individual- and sex-specific genotyping to be applicable to a diversity of mammalian species.

QR code technology in individual identification training provides an introduction in hands-on forensic DNA genotyping to medical students.

BACKGROUND: Forensic biology is a subject in the field of forensic science that stresses practical teaching and training in laboratory skills. Visualization of deoxyribonucleic acid (DNA) profiles is important in individual identification and is easily performed by well-trained examiners. Therefore, developing a novel training project for obtaining individual DNA profiles can improve the quality of teaching for medical students or trainees. DNA profiles based on quick response (QR) codes can also be applied to practical teaching and operation training for individual identification. METHODS: A novel training project was developed through an experimental course in forensic biology. Blood samples and buccal swabs with oral epithelial cells, as used in the forensic DNA laboratory, were obtained from medical students at Fujian Medical University. DNA was isolated, and a number of short tandem repeat (STR) loci were used as genetic markers to generate DNA profiles. The students converted DNA profiles and individual information into a QR code. The QR code could then be scanned by a mobile phone for consulting and retrieval. Gene identity cards with QR codes were produced and provided to every student. The participation rate and passing rate of students who participated in the novel training project were calculated and compared with those of students in the traditional experimental course, and a chi-square test was carried out by SPSS 23.0 software to evaluate the teaching effectiveness. p < 0.05 indicated significant differences. In addition, a survey was conducted to investigate the likelihood of using of gene identity cards with QR codes in the future. RESULTS: A total of 54 of 91 medical students who studied forensic biology participated in the novel training project in 2021. Only 31 of 78 students who studied forensic biology participated in the traditional experimental course in 2020. The participation rate in the novel training project was 24% higher than that of the traditional experimental course. The participants in the novel training project showed better performance in forensic biological handling techniques. The passing rate of the students in the forensic biology course with the novel training project was approximately 17% higher than that of the students in the former course. The participation rates and passing rates of the two groups were significantly different (χ = 6.452, p = 0.008 and χ = 11.043, p = 0.001). In the novel training project, all participants made 54 gene identity cards with QR codes. Furthermore, in the DNA profiles of four African students who participated, we found two rare alleles that were not discovered in Asians. The survey showed that the use of gene identity cards with QR codes was accepted by most participants, and the likelihood of future utilization was 78%. CONCLUSION: We established a novel training project to promote the learning activities of medical students in experimental forensic biology courses. The participants showed great interest in using gene identity cards with QR codes to store general individual identity information and DNA profiles. They also examined the genetic population differences between different races based on DNA profiles. Hence, the novel training project could be useful for training workshops, forensic experimental courses, and medical big data research.

Convolutional Neural Network for Individual Identification Using Phase Space Reconstruction of Electrocardiogram.

Electrocardiogram (ECG) biometric provides an authentication to identify an individual on the basis of specific cardiac potential measured from a living body. Convolutional neural networks (CNN) outperform traditional ECG biometrics because convolutions can produce discernible features from ECG through machine learning. Phase space reconstruction (PSR), using a time delay technique, is one of the transformations from ECG to a feature map, without the need of exact R-peak alignment. However, the effects of time delay and grid partition on identification performance have not been investigated. In this study, we developed a PSR-based CNN for ECG biometric authentication and examined the aforementioned effects. Based on a population of 115 subjects selected from the PTB Diagnostic ECG Database, a higher identification accuracy was achieved when the time delay was set from 20 to 28 ms, since it produced a well phase-space expansion of P, QRS, and T waves. A higher accuracy was also achieved when a high-density grid partition was used, since it produced a fine-detail phase-space trajectory. The use of a scaled-down network for PSR over a low-density grid with 32 × 32 partitions achieved a comparable accuracy with using a large-scale network for PSR over 256 × 256 partitions, but it had the benefit of reductions in network size and training time by 10 and 5 folds, respectively.

Individual Pig Identification Using Back Surface Point Clouds in 3D Vision.

The individual identification of pigs is the basis for precision livestock farming (PLF), which can provide prerequisites for personalized feeding, disease monitoring, growth condition monitoring and behavior identification. Pig face recognition has the problem that pig face samples are difficult to collect and images are easily affected by the environment and body dirt. Due to this problem, we proposed a method for individual pig identification using three-dimension (3D) point clouds of the pig's back surface. Firstly, a point cloud segmentation model based on the PointNet++ algorithm is established to segment the pig's back point clouds from the complex background and use it as the input for individual recognition. Then, an individual pig recognition model based on the improved PointNet++LGG algorithm was constructed by increasing the adaptive global sampling radius, deepening the network structure and increasing the number of features to extract higher-dimensional features for accurate recognition of different individuals with similar body sizes. In total, 10,574 3D point cloud images of ten pigs were collected to construct the dataset. The experimental results showed that the accuracy of the individual pig identification model based on the PointNet++LGG algorithm reached 95.26%, which was 2.18%, 16.76% and 17.19% higher compared with the PointNet model, PointNet++SSG model and MSG model, respectively. Individual pig identification based on 3D point clouds of the back surface is effective. This approach is easy to integrate with functions such as body condition assessment and behavior recognition, and is conducive to the development of precision livestock farming.

Process of Forensic Medicine in DNA Identification of Aged Human Remains.

Skeleton and teeth are important biological samples. Due to their special structure and strong ability to resist degradation, they are ideal biological materials to retain DNA under natural condition. In many cases, such as historical figure identification, aged skeleton and teeth are usually the only biological samples. However, their DNA is in a state of trace, damage and degradation to different degrees, which requires special experimental treatment to achieve identification. This paper reviews the sample selection, DNA extraction, DNA enrichment and analysis approaches based on relevant research reports in recent years, aiming to promote the further development and improvement of the aged skeleton and teeth identification system.

Forensic Validation and Application Evaluation of IDentifier DNA Typing Kit (Yan-Huang34).

OBJECTIVES: To investigate the technical performance of IDentifier DNA typing kit (YanHuang34) and evaluate its forensic application value. METHODS: Following the Criterion of Forensic Science Human Fluorescence STR Multiplex Amplification Reagent (GB/T 37226-2018), IDentifier DNA typing kit (YanHuang34) was verified in 11 aspects of species specificity, veracity, sensibility, adaptability, inhibitor tolerance, consistency, balance, reaction condition verification, mixed samples, stability and inter batch consistency. The system efficiency of IDentifier DNA typing kit (YanHuang34) was compared with the PowerPlex® Fusion 6C System, VersaPlex® 27PY System and VeriFilerTM Plus PCR Amplification Kit. The IDentifier DNA typing kit (YanHuang34) was used to detect the swabs of biological samples in daily cases and the STR performances were observed. RESULTS: IDentifier DNA typing kit (YanHuang34) had good species specificity, veracity, adaptability, inhibitor tolerance and balance. The sensibility was up to 0.062 5 ng. It was able to detect different types of samples, degraded samples and inhibitor mixed samples. Complete DNA typing could be obtained for samples with the mixture ratio less than 4∶1. The system efficiency of IDentifier DNA typing kit (YanHuang34) was very high, with TDP up to 1-1.08×10-37, CPEtrio and CPEduo up to 1-5.47×10-14 and 1-6.43×10-9, respectively. For the touched biological samples in actual cases, the effective detection rate was 21.05%. The system efficiency of kinship, single parent and full sibling identifications was effectively improved. CONCLUSIONS: The IDentifier DNA typing kit (YanHuang34) is adaptive to the GB/T 37226-2018 requirements. It can be used for individual identification and paternity identification, and is suitable for application in the field of forensic science.

Robustness of individualized inferences from longitudinal resting state EEG dynamics.

Tracking how individual human brains change over extended timescales is crucial to clinical scenarios ranging from stroke recovery to healthy aging. The use of resting state (RS) activity for tracking is a promising possibility. However, it is unresolved how a person's RS activity over time can be decoded to distinguish neurophysiological changes from confounding cognitive variability. Here, we develop a method to screen RS activity changes for these confounding effects by formulating it as a problem of change classification. We demonstrate a novel solution to change classification by linking individual-specific change to inter-individual differences. Individual RS-electroencephalography (EEG) was acquired over 5 consecutive days including task states devised to simulate the effects of inter-day cognitive variation. As inter-individual differences are shaped by neurophysiological differences, the inter-individual differences in RS activity on 1 day were analysed (using machine learning) to identify distinctive configurations in each individual's RS activity. Using this configuration as a decision rule, an individual could be re-identified from 2-s samples of the instantaneous oscillatory power spectrum acquired on a different day both from RS and confounded RS with a limited loss in accuracy. Importantly, the low loss in accuracy in cross-day versus same-day classification was achieved with classifiers that combined information from multiple frequency bands at channels across the scalp (with a concentration at characteristic fronto-central and occipital zones). Taken together, these findings support the technical feasibility of screening RS activity for confounding effects and the suitability of longitudinal RS for robust individualized inferences about neurophysiological change in health and disease.

Forensic efficacy evaluation and genetic structure exploration of the Yunnan Miao group by a multiplex InDel panel.

The aim of the study was to better understand the genetic characteristics of the Miao group in China. Herein, genetic characteristics and forensic application values of 57 autosomal insertion-deletion (InDel) loci were investigated in 210 unrelated healthy individuals from the Chinese Yunnan Miao (YM) group. Meanwhile, the genetic differences in these InDels were compared among the YM group and 26 reference populations. The results of forensic statistical analyses showed that all 57 autosomal InDels were in accordance with the Hardy-Weinberg and linkage equilibria of pairwise loci in the Chinese YM group. Moreover, the combined probability of discrimination and probability of exclusion in the YM group were 0.9999999999999999999999801 and 0.999928, respectively, which indicated that the multiplex amplification including 57 autosomal InDels was suitable for forensic individual identification and paternity testing in the Chinese YM group. In addition, the results of allelic frequency distribution differential analyses, principal component analyses, phylogenetic tree reconstruction, and genetic structure analyses between the Chinese YM group and 26 reference populations revealed that the genetic similarities between the YM group and East Asian populations were more than that between the YM group and other geographical populations. This 57 autosomal InDels system can also effectively distinguish East Asian, European, and African populations.

Detection of deletion/insertion polymorphism profiles from single human hair shafts.

BACKGROUND: Hair is a frequently encountered biological evidence in personal identification. The amount of nuclear DNA that can be extracted from a single strand of rootless hair is most limited, making the detection of short tandem repeat (STR) polymorphisms difficult. To overcome these limitations, deletion/insertion polymorphisms (DIP) as a new type of genetic marker have shown their benefits in detecting low-copy-number DNA. The Investigator DIPplex kit contains 30 biallelic autosomal DIP and amelogenin. The analysis of DIPs combines the advantages of both STR and single nucleotide polymorphism analyses. Thus, this study aimed to detect the DIP distribution of individual hair shafts from individuals. METHODS AND RESULTS: DNA was extracted from the shaft of fresh, aged, and shed hair. After DNA was evaluated, the DIP profiles were detected by capillary electrophoresis. The results indicated that the amount of DNA extracted from hair roots was much higher than that from the hair shafts in the same individual for all samples. The degradation index values of DNA from the aged hair shafts were highest. It is classified to be "mildly degraded." Compared with their hair roots, the full DIP profiles were detected for fresh hair, 70% for aged hair, and 92% for shed hair. Contrarily, except for fresh hair shafts, only three STR loci of the aged and shed strands of hair could be genotyped using AmpFlSTR MiniFiler PCR Amplification Kit. CONCLUSIONS: These results indicate that the detection of DIP profile is an effective tool for personal identification from hair shafts, including aged hair.

Individual Identification by Late Information Fusion of EmgCNN and EmgLSTM from Electromyogram Signals.

This paper is concerned with individual identification by late fusion of two-stream deep networks from Electromyogram (EMG) signals. EMG signal has more advantages on security compared to other biosignals exposed visually, such as the face, iris, and fingerprints, when used for biometrics, at least in the aspect of visual exposure, because it is measured through contact without any visual exposure. Thus, we propose an ensemble deep learning model by late information fusion of convolutional neural networks (CNN) and long short-term memory (LSTM) from EMG signals for robust and discriminative biometrics. For this purpose, in the ensemble model's first stream, one-dimensional EMG signals were converted into time-frequency representation to train a two-dimensional convolutional neural network (EmgCNN). In the second stream, statistical features were extracted from one-dimensional EMG signals to train a long short-term memory (EmgLSTM) that uses sequence input. Here, the EMG signals were divided into fixed lengths, and feature values were calculated for each interval. A late information fusion is performed by the output scores of two deep learning models to obtain a final classification result. To confirm the superiority of the proposed method, we use an EMG database constructed at Chosun University and a public EMG database. The experimental results revealed that the proposed method showed performance improvement by 10.76% on average compared to a single stream and the previous methods.

Real-Time Prediction of Growth Characteristics for Individual Fruits Using Deep Learning.

Understanding the growth status of fruits can enable precise growth management and improve the product quality. Previous studies have rarely used deep learning to observe changes over time, and manual annotation is required to detect hidden regions of fruit. Thus, additional research is required for automatic annotation and tracking fruit changes over time. We propose a system to record the growth characteristics of individual apples in real time using Mask R-CNN. To accurately detect fruit regions hidden behind leaves and other fruits, we developed a region detection model by automatically generating 3000 composite orchard images using cropped images of leaves and fruits. The effectiveness of the proposed method was verified on a total of 1417 orchard images obtained from the monitoring system, tracking the size of fruits in the images. The mean absolute percentage error between the true value manually annotated from the images and detection value provided by the proposed method was less than 0.079, suggesting that the proposed method could extract fruit sizes in real time with high accuracy. Moreover, each prediction could capture a relative growth curve that closely matched the actual curve after approximately 150 elapsed days, even if a target fruit was partially hidden.

The value of individual identification in studies of free-living hyenas and aardwolves.

From population estimates to social evolution, much of our understanding of the family Hyaenidae is drawn from studies of known individuals. The extant species in this family (spotted hyenas, Crocuta crocuta, brown hyenas, Parahyaena brunnea, striped hyenas, Hyaena hyaena, and aardwolves, Proteles cristata) are behaviorally diverse, presenting an equally diverse set of logistical constraints on capturing and marking individuals. All these species are individually identifiable by their coat patterns, providing a useful alternative to man-made markings. Many studies have demonstrated the utility of this method in answering a wide range of research questions across all four species, with some employing a creative fusion of techniques. Despite its pervasiveness in basic research on hyenas and aardwolves, individual identification has rarely been applied to the conservation and management of these species. We argue that individual identification using naturally occurring markings in applied research could prove immensely helpful, as this could further improve accuracy of density estimates, reveal characteristics of suitable habitat, identify threats to population persistence, and help to identify individual problem animals. Supplementary Information: The online version contains supplementary material available at 10.1007/s42991-022-00309-4.