Application of deep learning with bivariate models for genomic prediction of sow lifetime productivity-related traits.

OBJECTIVE: Pig breeders cannot obtain phenotypic information at the time of selection for sow lifetime productivity (SLP). They would benefit from obtaining genetic information of candidate sows. Genomic data interpreted using deep learning (DL) techniques could contribute to the genetic improvement of SLP to maximize farm profitability because DL models capture nonlinear genetic effects such as dominance and epistasis more efficiently than conventional genomic prediction methods based on linear models. This study aimed to investigate the usefulness of DL for the genomic prediction of two SLP-related traits; lifetime number of litters (LNL) and lifetime pig production (LPP). METHODS: Two bivariate DL models, convolutional neural network (CNN) and local convolutional neural network (LCNN), were compared with conventional bivariate linear models (i.e., genomic best linear unbiased prediction, Bayesian ridge regression, Bayes A, and Bayes B). Phenotype and pedigree data were collected from 40,011 sows that had husbandry records. Among these, 3,652 pigs were genotyped using the PorcineSNP60K BeadChip. RESULTS: The best predictive correlation for LNL was obtained with CNN (0.28), followed by LCNN (0.26) and conventional linear models (approximately 0.21). For LPP, the best predictive correlation was also obtained with CNN (0.29), followed by LCNN (0.27) and conventional linear models (approximately 0.25). A similar trend was observed with the mean squared error of prediction for the SLP traits. CONCLUSION: This study provides an example of a CNN that can outperform against the linear model-based genomic prediction approaches when the nonlinear interaction components are important because LNL and LPP exhibited strong epistatic interaction components. Additionally, our results suggest that applying bivariate DL models could also contribute to the prediction accuracy by utilizing the genetic correlation between LNL and LPP.

In-Home Smartphone-Based Prediction of Obstructive Sleep Apnea in Conjunction With Level 2 Home Polysomnography.

Importance: Consumer-level sleep analysis technologies have the potential to revolutionize the screening for obstructive sleep apnea (OSA). However, assessment of OSA prediction models based on in-home recording data is usually performed concurrently with level 1 in-laboratory polysomnography (PSG). Establishing the predictability of OSA using sound data recorded from smartphones based on level 2 PSG at home is important. Objective: To validate the performance of a prediction model for OSA using breathing sound recorded from smartphones in conjunction with level 2 PSG at home. Design, Setting, and Participants: This diagnostic study followed a prospective design, involving participants who underwent unattended level 2 home PSG. Breathing sounds were recorded during sleep using 2 smartphones, one with an iOS operating system and the other with an Android operating system, simultaneously with home PSG in participants' own home environment. Participants were 19 years and older, slept alone, and had either been diagnosed with OSA or had no previous diagnosis. The study was performed between February 2022 and February 2023. Main Outcomes and Measures: Sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of the predictive model based on the recorded breathing sounds. Results: Of the 101 participants included during the study duration, the mean (SD) age was 48.3 (14.9) years, and 51 (50.5%) were female. For the iOS smartphone, the sensitivity values at apnea-hypopnea index (AHI) levels of 5, 15, and 30 per hour were 92.6%, 90.9%, and 93.3%, respectively, with specificities of 84.3%, 94.4%, and 94.4%, respectively. Similarly, for the Android smartphone, the sensitivity values at AHI levels of 5, 15, and 30 per hour were 92.2%, 90.0%, and 92.9%, respectively, with specificities of 84.0%, 94.4%, and 94.3%, respectively. The accuracy for the iOS smartphone was 88.6%, 93.3%, and 94.3%, respectively, and for the Android smartphone was 88.1%, 93.1%, and 94.1% at AHI levels of 5, 15, and 30 per hour, respectively. Conclusions and Relevance: This diagnostic study demonstrated the feasibility of predicting OSA with a reasonable level of accuracy using breathing sounds obtained by smartphones during sleep at home.

Accuracy of 11 Wearable, Nearable, and Airable Consumer Sleep Trackers: Prospective Multicenter Validation Study.

BACKGROUND: Consumer sleep trackers (CSTs) have gained significant popularity because they enable individuals to conveniently monitor and analyze their sleep. However, limited studies have comprehensively validated the performance of widely used CSTs. Our study therefore investigated popular CSTs based on various biosignals and algorithms by assessing the agreement with polysomnography. OBJECTIVE: This study aimed to validate the accuracy of various types of CSTs through a comparison with in-lab polysomnography. Additionally, by including widely used CSTs and conducting a multicenter study with a large sample size, this study seeks to provide comprehensive insights into the performance and applicability of these CSTs for sleep monitoring in a hospital environment. METHODS: The study analyzed 11 commercially available CSTs, including 5 wearables (Google Pixel Watch, Galaxy Watch 5, Fitbit Sense 2, Apple Watch 8, and Oura Ring 3), 3 nearables (Withings Sleep Tracking Mat, Google Nest Hub 2, and Amazon Halo Rise), and 3 airables (SleepRoutine, SleepScore, and Pillow). The 11 CSTs were divided into 2 groups, ensuring maximum inclusion while avoiding interference between the CSTs within each group. Each group (comprising 8 CSTs) was also compared via polysomnography. RESULTS: The study enrolled 75 participants from a tertiary hospital and a primary sleep-specialized clinic in Korea. Across the 2 centers, we collected a total of 3890 hours of sleep sessions based on 11 CSTs, along with 543 hours of polysomnography recordings. Each CST sleep recording covered an average of 353 hours. We analyzed a total of 349,114 epochs from the 11 CSTs compared with polysomnography, where epoch-by-epoch agreement in sleep stage classification showed substantial performance variation. More specifically, the highest macro F1 score was 0.69, while the lowest macro F1 score was 0.26. Various sleep trackers exhibited diverse performances across sleep stages, with SleepRoutine excelling in the wake and rapid eye movement stages, and wearables like Google Pixel Watch and Fitbit Sense 2 showing superiority in the deep stage. There was a distinct trend in sleep measure estimation according to the type of device. Wearables showed high proportional bias in sleep efficiency, while nearables exhibited high proportional bias in sleep latency. Subgroup analyses of sleep trackers revealed variations in macro F1 scores based on factors, such as BMI, sleep efficiency, and apnea-hypopnea index, while the differences between male and female subgroups were minimal. CONCLUSIONS: Our study showed that among the 11 CSTs examined, specific CSTs showed substantial agreement with polysomnography, indicating their potential application in sleep monitoring, while other CSTs were partially consistent with polysomnography. This study offers insights into the strengths of CSTs within the 3 different classes for individuals interested in wellness who wish to understand and proactively manage their own sleep.

Prediction of Sleep Stages Via Deep Learning Using Smartphone Audio Recordings in Home Environments: Model Development and Validation.

BACKGROUND: The growing public interest and awareness regarding the significance of sleep is driving the demand for sleep monitoring at home. In addition to various commercially available wearable and nearable devices, sound-based sleep staging via deep learning is emerging as a decent alternative for their convenience and potential accuracy. However, sound-based sleep staging has only been studied using in-laboratory sound data. In real-world sleep environments (homes), there is abundant background noise, in contrast to quiet, controlled environments such as laboratories. The use of sound-based sleep staging at homes has not been investigated while it is essential for practical use on a daily basis. Challenges are the lack of and the expected huge expense of acquiring a sufficient size of home data annotated with sleep stages to train a large-scale neural network. OBJECTIVE: This study aims to develop and validate a deep learning method to perform sound-based sleep staging using audio recordings achieved from various uncontrolled home environments. METHODS: To overcome the limitation of lacking home data with known sleep stages, we adopted advanced training techniques and combined home data with hospital data. The training of the model consisted of 3 components: (1) the original supervised learning using 812 pairs of hospital polysomnography (PSG) and audio recordings, and the 2 newly adopted components; (2) transfer learning from hospital to home sounds by adding 829 smartphone audio recordings at home; and (3) consistency training using augmented hospital sound data. Augmented data were created by adding 8255 home noise data to hospital audio recordings. Besides, an independent test set was built by collecting 45 pairs of overnight PSG and smartphone audio recording at homes to examine the performance of the trained model. RESULTS: The accuracy of the model was 76.2% (63.4% for wake, 64.9% for rapid-eye movement [REM], and 83.6% for non-REM) for our test set. The macro F1-score and mean per-class sensitivity were 0.714 and 0.706, respectively. The performance was robust across demographic groups such as age, gender, BMI, or sleep apnea severity (accuracy 73.4%-79.4%). In the ablation study, we evaluated the contribution of each component. While the supervised learning alone achieved accuracy of 69.2% on home sound data, adding consistency training to the supervised learning helped increase the accuracy to a larger degree (+4.3%) than adding transfer learning (+0.1%). The best performance was shown when both transfer learning and consistency training were adopted (+7.0%). CONCLUSIONS: This study shows that sound-based sleep staging is feasible for home use. By adopting 2 advanced techniques (transfer learning and consistency training) the deep learning model robustly predicts sleep stages using sounds recorded at various uncontrolled home environments, without using any special equipment but smartphones only.

Real-Time Detection of Sleep Apnea Based on Breathing Sounds and Prediction Reinforcement Using Home Noises: Algorithm Development and Validation.

BACKGROUND: Multinight monitoring can be helpful for the diagnosis and management of obstructive sleep apnea (OSA). For this purpose, it is necessary to be able to detect OSA in real time in a noisy home environment. Sound-based OSA assessment holds great potential since it can be integrated with smartphones to provide full noncontact monitoring of OSA at home. OBJECTIVE: The purpose of this study is to develop a predictive model that can detect OSA in real time, even in a home environment where various noises exist. METHODS: This study included 1018 polysomnography (PSG) audio data sets, 297 smartphone audio data sets synced with PSG, and a home noise data set containing 22,500 noises to train the model to predict breathing events, such as apneas and hypopneas, based on breathing sounds that occur during sleep. The whole breathing sound of each night was divided into 30-second epochs and labeled as "apnea," "hypopnea," or "no-event," and the home noises were used to make the model robust to a noisy home environment. The performance of the prediction model was assessed using epoch-by-epoch prediction accuracy and OSA severity classification based on the apnea-hypopnea index (AHI). RESULTS: Epoch-by-epoch OSA event detection showed an accuracy of 86% and a macro F1-score of 0.75 for the 3-class OSA event detection task. The model had an accuracy of 92% for "no-event," 84% for "apnea," and 51% for "hypopnea." Most misclassifications were made for "hypopnea," with 15% and 34% of "hypopnea" being wrongly predicted as "apnea" and "no-event," respectively. The sensitivity and specificity of the OSA severity classification (AHI≥15) were 0.85 and 0.84, respectively. CONCLUSIONS: Our study presents a real-time epoch-by-epoch OSA detector that works in a variety of noisy home environments. Based on this, additional research is needed to verify the usefulness of various multinight monitoring and real-time diagnostic technologies in the home environment.

Development of a Pedicle Screw Fixation Simulation Model for Surgical Training Using a 3-Dimensional Printer.

OBJECTIVE: Training surgeons in pedicle screw fixation (PSF) techniques during actual surgery is limited because of patient safety, complications, and surgical efficiency issues. Recent technical developments are leading the world to an era of personalized three-dimensional (3D) printing. This study aimed to evaluate the educational effect of using a 3D-printed spine model to train beginners in PSF techniques to improve screw accuracy and procedure time. METHODS: Computed tomography (CT) scan data were used in a 3D printer to produce a life-size lumbar spine replica of L1-3 vertebrae. Four residents performed PSF thrice. Each resident performed 18 screw fixations on both sides (6 screws per trial). The time to complete the procedure and pedicle violation was recorded. RESULTS: The average time for the 3 procedures was 42.1±2.9 minutes, 38.8±3.3 minutes, and 32.1±2.5 minutes, respectively. Furthermore, the average pedicle screw score for the 3 procedures was 13.0±0.8, 14.5±0.6, and 16.0±0.8, respectively. As the trial was repeated, the procedure time decreased and the accuracy of screw fixation tended to be more accurate. CONCLUSIONS: It was possible to decrease the procedure time and increase accuracy through repeated training using the 3D-printed spine model. By implementing a 3Dprinted spine model based on the patient's actual CT data, surgeons can perform simulation surgery before the actual surgery. Therefore, this technology can be useful in educating residents to improve their surgical skills.

Genetic Parameter Estimation and Genome-Wide Association Analysis of Social Genetic Effects on Average Daily Gain in Purebreds and Crossbreds.

Average daily gain (ADG) is an important growth trait in the pig industry. The direct genetic effect (DGE) has been studied mainly to assess the association between genetic information and economic traits. The social genetic effect (SGE) has been shown to affect ADG simultaneously with the DGE because of group housing systems. We conducted this study to elucidate the genetic characteristics and relationships of the DGE and SGE of purebred Korean Duroc and crossbred pigs by single-step genomic best linear unbiased prediction and a genome-wide association study. We used the genotype, phenotype, and pedigree data of 1779, 6022, and 7904 animals, respectively. Total heritabilities on ADG were 0.19 ± 0.04 and 0.39 ± 0.08 for purebred and crossbred pigs, respectively. The genetic correlation was the greatest (0.77 ± 0.12) between the SGE of purebred and DGE of crossbred pigs. We found candidate genes located in the quantitative trait loci (QTLs) for the SGE that were associated with behavior and neurodegenerative diseases, and candidate genes in the QTLs for DGE that were related to body mass, size of muscle fiber, and muscle hypertrophy. These results suggest that the genomic selection of purebred animals could be applied for crossbred performance.

End-to-End Sleep Staging Using Nocturnal Sounds from Microphone Chips for Mobile Devices.

Purpose: Nocturnal sounds contain numerous information and are easily obtainable by a non-contact manner. Sleep staging using nocturnal sounds recorded from common mobile devices may allow daily at-home sleep tracking. The objective of this study is to introduce an end-to-end (sound-to-sleep stages) deep learning model for sound-based sleep staging designed to work with audio from microphone chips, which are essential in mobile devices such as modern smartphones. Patients and Methods: Two different audio datasets were used: audio data routinely recorded by a solitary microphone chip during polysomnography (PSG dataset, N=1154) and audio data recorded by a smartphone (smartphone dataset, N=327). The audio was converted into Mel spectrogram to detect latent temporal frequency patterns of breathing and body movement from ambient noise. The proposed neural network model learns to first extract features from each 30-second epoch and then analyze inter-epoch relationships of extracted features to finally classify the epochs into sleep stages. Results: Our model achieved 70% epoch-by-epoch agreement for 4-class (wake, light, deep, REM) sleep stage classification and robust performance across various signal-to-noise conditions. The model performance was not considerably affected by sleep apnea or periodic limb movement. External validation with smartphone dataset also showed 68% epoch-by-epoch agreement. Conclusion: The proposed end-to-end deep learning model shows potential of low-quality sounds recorded from microphone chips to be utilized for sleep staging. Future study using nocturnal sounds recorded from mobile devices at home environment may further confirm the use of mobile device recording as an at-home sleep tracker.

Genome-Wide Assessment of a Korean Composite Pig Breed, Woori-Heukdon.

A Korean synthetic pig breed, Woori-Heukdon (WRH; F3), was developed by crossing parental breeds (Korean native pig [KNP] and Korean Duroc [DUC]) with their crossbred populations (F1 and F2). This study in genome-wide assessed a total of 2,074 pigs which include the crossbred and the parental populations using the Illumina PorcineSNP60 BeadChip. After quality control of the initial datasets, we performed population structure, genetic diversity, and runs of homozygosity (ROH) analyses. Population structure analyses showed that crossbred populations were genetically influenced by the parental breeds according to their generation stage in the crossbreeding scheme. Moreover, principal component analysis showed the dispersed cluster of WRH, which might reflect introducing a new breeding group into the previous one. Expected heterozygosity values, which were used to assess genetic diversity, were .365, .349, .336, .330, and .211 for WRH, F2, F1, DUC, and KNP, respectively. The inbreeding coefficient based on ROH was the highest in KNP (.409), followed by WRH (.186), DUC (.178), F2 (.107), and F1 (.035). Moreover, the frequency of short ROH decreased according to the crossing stage (from F1 to WRH). Alternatively, the frequency of medium and long ROH increased, which indicated recent inbreeding in F2 and WRH. Furthermore, gene annotation of the ROH islands in WRH that might be inherited from their parental breeds revealed several interesting candidate genes that may be associated with adaptation, meat quality, production, and reproduction traits in pigs.

Association of the ubiquitin specific peptidase 9X -linked and Afadin expression patterns with sexual maturation in boar testis.

Closely correlated expression patterns between ubiquitin specific peptidase 9X-linked (USP9X) and adherens junction formation factor (Afadin) in mouse testis development suggests that Usp9x regulates the deubiquitination of Af-6 (also known as Afadin, AFDN), and subsequently, the cell adhesion dynamics during gametogenesis. However, this relationship has not yet been tested in other domestic animals. The study was examined the temporal and spatial expression patterns of porcine USP9X and AFDN from the pre-pubertal to adult stages using real time-PCR and immunohistochemistry. Furthermore, we detected the transcripts of USP9X and AFDN in the testis of 1-, 6- and 12-months old boar, respectively. USP9X and AFDN were found to have similar expressions patterns, with basal expression after 1 month followed by a significant up-regulation from 6 months (puberty) onwards. In addition, neither the AFDN or USP9X proteins were detected in spermatogenic cells but they were expressed in the leydig cells and sertoli cells. USP9X was detected around the basal lamina during pre-puberty, and predominantly expressed in the leydig cells at puberty. Finally, in adult testis, USP9X was increased at the sertoli cell-cell interface and the sertoli cell-spermatid interface. In summary, closely correlated expression patterns between USP9X and AFDN in boar testis supports the previous findings in mice. Furthermore, the junction connections between the sertoli cells may be regulated by the ubiquitination process mediated via USP9X.

Confidence-Based Framework Using Deep Learning for Automated Sleep Stage Scoring.

STUDY OBJECTIVES: Automated sleep stage scoring is not yet vigorously used in practice because of the black-box nature and the risk of wrong predictions. The objective of this study was to introduce a confidence-based framework to detect the possibly wrong predictions that would inform clinicians about which epochs would require a manual review and investigate the potential to improve accuracy for automated sleep stage scoring. METHODS: We used 702 polysomnography studies from a local clinical dataset (SNUBH dataset) and 2804 from an open dataset (SHHS dataset) for experiments. We adapted the state-of-the-art TinySleepNet architecture to train the classifier and modified the ConfidNet architecture to train an auxiliary confidence model. For the confidence model, we developed a novel method, Dropout Correct Rate (DCR), and the performance of it was compared with other existing methods. RESULTS: Confidence estimates (0.754) reflected accuracy (0.758) well in general. The best performance for differentiating correct and wrong predictions was shown when using the DCR method (AUROC: 0.812) compared to the existing approaches which largely failed to detect wrong predictions. By reviewing only 20% of epochs that received the lowest confidence values, the overall accuracy of sleep stage scoring was improved from 76% to 87%. For patients with reduced accuracy (ie, individuals with obesity or severe sleep apnea), the possible improvement range after applying confidence estimation was even greater. CONCLUSION: To the best of our knowledge, this is the first study applying confidence estimation on automated sleep stage scoring. Reliable confidence estimates by the DCR method help screen out most of the wrong predictions, which would increase the reliability and interpretability of automated sleep stage scoring.

Mechanical Properties and Metal-Ceramic Bond Strength of Co-Cr Alloy Manufactured by Selective Laser Melting.

Cobalt-chromium (Co-Cr) metal is one of the widely used biomaterials in the fabrication of dental prosthesis. The purpose of this study was to investigate whether there are differences in the properties of metals and bond strength with ceramics depending on the manufacturing methods of Co-Cr alloy. Co-Cr alloy specimens were prepared in three different ways: casting, milling, and selective laser melting (SLM). The mechanical properties (elastic modulus, yield strength, and flexural strength) of the alloys were investigated by flexure method in three-point bending mode, and microstructures of the specimens were analyzed. After application of the veneering ceramic through the three-point bending test, bond strength of the Metal-Ceramic was investigated. The cracked surfaces were observed by means of energy dispersive X-ray (EDX) spectroscopy and scanning electron microscopy (SEM) with backscattered electron (BSE) images. In mechanical properties, the elastic modulus was highest for the casting group, and the yield strength and flexural strength were lowest for the milling group. The SLM group showed finer homogeneous crystalline-microstructure, and a layered structure was observed at the fractured surface. After the ceramic bond strength test, all groups showed a mixed failure pattern. The casting group showed the highest bond strengths, whereas there was no significant difference between the other two groups. However, all groups have met the standard of bond strength according to international standards organization (ISO) with the appropriate passing rate. The results of this study indicate that the SLM manufacturing method may have the potential to replace traditional techniques for fabricating dental prosthesis.

Genome-wide investigation of a Korean synthetic breed, Woori-Heukdon using the Illumina PorcineSNP60K BeadChip.

BACKGROUND: Woori-Heukdon (KWH) is a Korean synthetic pig breed generated using Chookjin-Duroc (KCD), Chookjin-Chamdon (KCC), and their crossbreds. Currently, there is a severe lack of studies investigating the Korean breed populations including wild boars (KWB) throughout the genome. OBJECTIVE: This study was performed to investigate the genetic characteristics of Korean pig populations at the genome-wide level. METHODS: Using the SNP dataset derived from genotyped and downloaded datasets using the Illumina PorcineSNP60K BeadChip, we compared the genomes of 532 individuals derived from 23 pig breeds to assess the genetic diversity, inbreeding coefficient, genetic differentiation, and population structure. RESULTS: KWB showed the lowest average expected heterozygosity (HE = 0.1904), while KWH showed the highest genetic diversity (HE = 0.02859) among Korean populations. We verified that the genetic composition of KWH, showing USD of 74.8% and KCC of 25.2% in ADMIXTURE analysis. In population structure analyses, KCC was consistently shown to be separated from other pig populations. In addition, we observed gene flow from Western pigs to a part of Chinese populations. CONCLUSION: This study showed that Korean native pigs, KCC have genetic differences in comparison with Chinese and Western pigs; despite some historical records and recent genetic studies, we could not find any clear evidence that KCC was significantly influenced by Chinese or Western breeds in this study. We also verified the theoretical genomic composition of KWH at the molecular level in structure analyses. To our knowledge, this is the first genomic study to investigate the genomic characteristics of KWH and KCC.

Single-step genome-wide association study for social genetic effects and direct genetic effects on growth in Landrace pigs.

In livestock social interactions, social genetic effects (SGE) represent associations between phenotype of one individual and genotype of another. Such associations occur when the trait of interest is affected by transmissible phenotypes of social partners. The aim of this study was to estimate SGE and direct genetic effects (DGE, genetic effects of an individual on its own phenotype) on average daily gain (ADG) in Landrace pigs, and to conduct single-step genome-wide association study using SGE and DGE as dependent variables to identify quantitative trait loci (QTLs) and their positional candidate genes. A total of 1,041 Landrace pigs were genotyped using the Porcine SNP 60K BeadChip. Estimates of the two effects were obtained using an extended animal model. The SGE contributed 16% of the total heritable variation of ADG. The total heritability estimated by the extended animal model including both SGE and DGE was 0.52. The single-step genome-wide association study identified a total of 23 QTL windows for the SGE on ADG distributed across three chromosomes (i.e., SSC1, SSC2, and SSC6). Positional candidate genes within these QTL regions included PRDM13, MAP3K7, CNR1, HTR1E, IL4, IL5, IL13, KIF3A, EFHD2, SLC38A7, mTOR, CNOT1, PLCB2, GABRR1, and GABRR2, which have biological roles in neuropsychiatric processes. The results of biological pathway and gene network analyses also support the association of the neuropsychiatric processes with SGE on ADG in pigs. Additionally, a total of 11 QTL windows for DGE on ADG in SSC2, 3, 6, 9, 10, 12, 14, 16, and 17 were detected with positional candidate genes such as ARL15. We found a putative pleotropic QTL for both SGE and DGE on ADG on SSC6. Our results in this study provide important insights that can help facilitate a better understanding of the molecular basis of SGE for socially affected traits.

Detection of Atrial Fibrillation Using a Ring-Type Wearable Device (CardioTracker) and Deep Learning Analysis of Photoplethysmography Signals: Prospective Observational Proof-of-Concept Study.

BACKGROUND: Continuous photoplethysmography (PPG) monitoring with a wearable device may aid the early detection of atrial fibrillation (AF). OBJECTIVE: We aimed to evaluate the diagnostic performance of a ring-type wearable device (CardioTracker, CART), which can detect AF using deep learning analysis of PPG signals. METHODS: Patients with persistent AF who underwent cardioversion were recruited prospectively. We recorded PPG signals at the finger with CART and a conventional pulse oximeter before and after cardioversion over a period of 15 min (each instrument). Cardiologists validated the PPG rhythms with simultaneous single-lead electrocardiography. The PPG data were transmitted to a smartphone wirelessly and analyzed with a deep learning algorithm. We also validated the deep learning algorithm in 20 healthy subjects with sinus rhythm (SR). RESULTS: In 100 study participants, CART generated a total of 13,038 30-s PPG samples (5850 for SR and 7188 for AF). Using the deep learning algorithm, the diagnostic accuracy, sensitivity, specificity, positive-predictive value, and negative-predictive value were 96.9%, 99.0%, 94.3%, 95.6%, and 98.7%, respectively. Although the diagnostic accuracy decreased with shorter sample lengths, the accuracy was maintained at 94.7% with 10-s measurements. For SR, the specificity decreased with higher variability of peak-to-peak intervals. However, for AF, CART maintained consistent sensitivity regardless of variability. Pulse rates had a lower impact on sensitivity than on specificity. The performance of CART was comparable to that of the conventional device when using a proper threshold. External validation showed that 94.99% (16,529/17,400) of the PPG samples from the control group were correctly identified with SR. CONCLUSIONS: A ring-type wearable device with deep learning analysis of PPG signals could accurately diagnose AF without relying on electrocardiography. With this device, continuous monitoring for AF may be promising in high-risk populations. TRIAL REGISTRATION: ClinicalTrials.gov NCT04023188; https://clinicaltrials.gov/ct2/show/NCT04023188.

Clinico-radiologic features of molar-incisor malformation in a case series of 38 patients: A retrospective observational study.

Molar-incisor malformation (MIM) is a recently defined dental abnormality of molar root and incisors, and introduced as one of the causes of periapical and periodontal abscess. The purpose of this study was to investigate the clinical and radiological features of MIM, with special emphasis on various medical history. A total of 38 patients with MIM were included in this study. Radiographic features and clinical data, including medical history, chief complaint, associated complications, treatment, and prognosis, were retrospectively investigated. On radiographs, the affected molars showed short, slender, underdeveloped roots and constricted pulp chambers. All affected incisors and canines exhibited dilacerated short roots, wedge-shaped defect on the cervical part of the crown. Complications included periodontal bone loss (52.6%), endodontic lesion (50.0%), and endodontic-periodontal lesion (28.9%). The medical histories of the patients with MIM indicate that almost all (94.7%) were hospitalized due to problems during the neonatal period. MIM may cause various dental problems, such as periapical and periodontal abscess and early loss of the affected teeth. The early diagnosis of MIM on radiographs and appropriate treatment will contribute to a favorable prognosis, especially for young and adolescent patients.

Deep Learning Approaches to Detect Atrial Fibrillation Using Photoplethysmographic Signals: Algorithms Development Study.

BACKGROUND: Wearable devices have evolved as screening tools for atrial fibrillation (AF). A photoplethysmographic (PPG) AF detection algorithm was developed and applied to a convenient smartphone-based device with good accuracy. However, patients with paroxysmal AF frequently exhibit premature atrial complexes (PACs), which result in poor unmanned AF detection, mainly because of rule-based or handcrafted machine learning techniques that are limited in terms of diagnostic accuracy and reliability. OBJECTIVE: This study aimed to develop deep learning (DL) classifiers using PPG data to detect AF from the sinus rhythm (SR) in the presence of PACs after successful cardioversion. METHODS: We examined 75 patients with AF who underwent successful elective direct-current cardioversion (DCC). Electrocardiogram and pulse oximetry data over a 15-min period were obtained before and after DCC and labeled as AF or SR. A 1-dimensional convolutional neural network (1D-CNN) and recurrent neural network (RNN) were chosen as the 2 DL architectures. The PAC indicator estimated the burden of PACs on the PPG dataset. We defined a metric called the confidence level (CL) of AF or SR diagnosis and compared the CLs of true and false diagnoses. We also compared the diagnostic performance of 1D-CNN and RNN with previously developed AF detectors (support vector machine with root-mean-square of successive difference of RR intervals and Shannon entropy, autocorrelation, and ensemble by combining 2 previous methods) using 10 5-fold cross-validation processes. RESULTS: Among the 14,298 training samples containing PPG data, 7157 samples were obtained during the post-DCC period. The PAC indicator estimated 29.79% (2132/7157) of post-DCC samples had PACs. The diagnostic accuracy of AF versus SR was 99.32% (70,925/71,410) versus 95.85% (68,602/71,570) in 1D-CNN and 98.27% (70,176/71,410) versus 96.04% (68,736/71,570) in RNN methods. The area under receiver operating characteristic curves of the 2 DL classifiers was 0.998 (95% CI 0.995-1.000) for 1D-CNN and 0.996 (95% CI 0.993-0.998) for RNN, which were significantly higher than other AF detectors (P<.001). If we assumed that the dataset could emulate a sufficient number of patients in training, both DL classifiers improved their diagnostic performances even further especially for the samples with a high burden of PACs. The average CLs for true versus false classification were 98.56% versus 78.75% for 1D-CNN and 98.37% versus 82.57% for RNN (P<.001 for all cases). CONCLUSIONS: New DL classifiers could detect AF using PPG monitoring signals with high diagnostic accuracy even with frequent PACs and could outperform previously developed AF detectors. Although diagnostic performance decreased as the burden of PACs increased, performance improved when samples from more patients were trained. Moreover, the reliability of the diagnosis could be indicated by the CL. Wearable devices sensing PPG signals with DL classifiers should be validated as tools to screen for AF.

Social behavior and group growth of finishing pigs with divergent social breeding values / Comportamiento social y crecimiento grupal de cerdos en etapa de finalización con valores divergentes de crianza sociales / Comportamento social e crescimento grupal de porcos de engorda com valores divergentes de criação social

Abstract Background: Behavioral traits of pigs have been shown to be partly under genetic control, which raises the possibility that behavior might be altered by genetic selection, resulting in pigs with better growth performance. Objective: To evaluate the behavior and growth of finishing pigs and investigate pigs selected for high or low social breeding value (SBV) in relation to social behavior and group growth. Methods: Thirty-five females and 35 boars from five positive and five negative SBV groups of finishing pigs were grown from 30 to 90 kg and housed in 10 test pens (3.0 × 3.3 m, 7 pigs/pen). Pigs were recorded with video technology for nine consecutive hours on days 1, 15, and 30 after mixing. Pigs were weighed at approximately 90 kg body weight and the number of days to reach 90 kg was then calculated. Results: The frequency and duration of behaviors were present in the positive and negative SBV groups after mixing. On day 1 after mixing, agonistic behavior was significantly higher (p=0.027) for the -SBV group compared with the +SBV group. Feeding and feeding-together behaviors were significantly higher (p<0.003) in the +SBV group on days 1 and 30 after mixing. Moreover, growth performance to reach 90 kg body weight was significantly faster (p<0.002) in the +SBV group than in the -SBV group. Conclusion: Social interactions, such as feeding-together behavior, among pen mates might affect their growth rate and feed intake. Selection for SBV could be used as an indirect technique for improving growth performance of pigs.

Resumen Antecedentes: Se ha demostrado que los rasgos conductuales de los cerdos están parcialmente bajo control genético, lo que plantea la posibilidad de que el comportamiento pueda ser alterado vía selección genética y resulte en cerdos con mejores rendimientos de crecimiento. Objetivo: Evaluar el comportamiento y crecimiento de los cerdos en etapa de finalización e investigar cerdos seleccionados por un valor alto o bajo de crianza social (SBV) en relación al comportamiento social y al crecimiento grupal. Métodos: Treinta y cinco hembras y 35 verracos, pertenecientes a cinco grupos positivos y cinco grupos negativos de SBV de cerdos en etapa de finalización, llevados hasta los 90, desde 30 kg de peso, alojados en 10 corrales de prueba (3,0 x 3,3 m, 7 cerdos/corral). Los cerdos fueron observados con la ayuda de tecnología de vídeo por nueve horas consecutivas en los días 1, 15 y 30 luego de ser mezclados. Además, los cerdos se pesaron a los 90 kg de peso aproximadamente y se calculó el número de días para alcanzar dicho peso. Resultados: La frecuencia y duración de los comportamientos de los cerdos en la etapa de finalización se presentaron en los grupos de SBV negativos y positivos luego de ser mezclados. El día 1 luego de la mezcla, el comportamiento agonístico fue significativamente mayor (p=0,027) en el grupo -SBV que en el grupo +SBV. Los comportamientos de consumo de alimento y de consumo en compañía fueron significativamente mayores (p<0,003) en el grupo +SBV en los días 1 y 30 luego de la mezcla. Además, el crecimiento para alcanzar 90 kg de peso corporal fue significativamente más rápido (p=0,002) en el grupo +SBV que el grupo -SBV. Conclusiones: Las interacciones sociales, tales como el comportamiento de consumo de alimento en compañía, entre los compañeros de corral, pueden afectar la tasa de crecimiento y consumo de alimento. La selección por SBV podría usarse como técnica indirecta para mejorar el rendimiento de crecimiento en cerdos.

Resumo Antecedentes: Os traços comportamentais dos porcos demonstraram estar parcialmente sob controle genético, o que aumenta a possibilidade de que o comportamento possa ser alterado pela seleção genética e resulte em porcos com melhor comportamento de crescimento. Objetivo: Avaliar o comportamento e o crescimento dos porcos de engorda e investigar os porcos selecionados para alto ou baixo valor de reprodução social (SBV) em relação ao comportamento social e crescimento do grupo. Métodos: Trinta e cinco fêmeas e 35 machos, pertencentes a cinco grupos de SBV positivos e cinco negativos de porcos de engorda, foram engordados até 90 de 30 kg e alojados em 10 currais de teste (3,0 × 3,3 m, 7 porcos/curral). Os porcos foram observados com o auxílio de tecnologia de vídeo durante nove horas consecutivas nos dias 1, 15 e 30 após a mistura. Além disso, os porcos foram sopesados em aproximadamente 90 kg de peso corporal e o número de dias para atingir 90 kg foi então calculado. Resultados: A frequência e a duração dos comportamentos dos porcos de engorda foram apresentadas com grupos de SBV positivo e negativo após a mistura. No dia 1 após a mistura, o comportamento agonístico foi significativamente maior (p=0,027) no grupo -SBV do que no grupo +SBV. Os comportamentos de alimentação e alimentação conjunta foram significativamente maiores (p<0,003) no grupo +SBV nos dias 1 e 30 após a mistura. Além disso, o comportamento de crescimento do grupo para atingir 90 kg de peso corporal foi significativamente mais rápido (p<0,002) no grupo +SBV do que no grupo -SBV. Conclusão: As interações sociais, como o comportamento de alimentação conjunta, entre companheiros de curral podem afetar a taxa de crescimento e a ingestão alimentar. A seleção para SBV pode ser uma técnica indireta para melhorar o comportamento de crescimento dos porcos.

Behaviors and body weight of suckling piglets in different social environments.

OBJECTIVE: This study was conducted to characterize the behaviors and the body weight of suckling piglets in different social environments. METHODS: Two groups of sows and suckling piglets housed either in individual farrowing crates in separate pens (1.8×2.4 m, the control group) or in groups of three sows with their piglets in farrowing crates in a large common enclosure (5.4×2.2 m, the treatment group) were observed with the aid of video technology for 9 consecutive hours on days 1, 2, and 3, after mixing. RESULTS: Suckling, agonistic, and elimination behaviors of suckling piglets were significantly higher in the control group than in the treatment group. Inactive behavior was higher in the treatment group than in the control group. Most of the effects of the social environment on the suckling piglets seem to be the result of large reductions in behaviors and body weight for piglets switching from high activity to low activity. Moreover, suckling behavior and birth body weight were highly correlated with body weight at the end of the test. CONCLUSION: The social environment that resulted from mixing, thus, had significant effects on the behavior and body weight of suckling piglets, and behavioral characteristics, therefore, should be considered when making improvements to the husbandry and care methods used in swine production.

Behavioral Characteristics of Weaned Piglets Mixed in Different Groups.

With regard to animal welfare concerns, behavioral information of weaned and mixed piglets is great interest in swine production. The aim of this study was to demonstrate the change in behavior of weaned piglets over time in two different groups (littermates and piglets from different litters) after mixing. Two weaned groups of piglets (72 individuals in all) housed either with littermates or with foreign piglets (6 individuals in 1.8 m×1.4 m pens, 28°C±1°C temperature) were observed with the aid of video technology for 9 consecutive hours on days 1, 2, and 3 after mixing. The behaviors of the weaned piglets in the control and treatment groups were significantly different among the days after mixing. Piglets were, however, more active and aggressive in the groups with foreign piglets. This study reveals a lower level of agonistic behavior in groups of piglets that came from the same litter.