Pulse rate estimation based on facial videos:an evaluation and optimization of the classical methods using both self-constructed and public datasets

Pulse rate is one of the important characteristics of traditional Chinese medicine pulse diagnosis,and it is of great significance for determining the nature of cold and heat in diseases.The prediction of pulse rate based on facial video is an exciting research field for getting palpation information by observation diagnosis.However,most studies focus on optimizing the algorithm based on a small sample of participants without systematically investigating multiple influencing factors.A total of 209 participants and 2,435 facial videos,based on our self-constructed Multi-Scene Sign Dataset and the public datasets,were used to perform a multi-level and multi-factor comprehensive comparison.The effects of different datasets,blood volume pulse signal extraction algorithms,region of interests,time windows,color spaces,pulse rate calculation methods,and video recording scenes were analyzed.Furthermore,we proposed a blood volume pulse signal quality optimization strategy based on the inverse Fourier transform and an improvement strategy for pulse rate estimation based on signal-to-noise ratio threshold sliding.We found that the effects of video estimation of pulse rate in the Multi-Scene Sign Dataset and Pulse Rate Detection Dataset were better than in other datasets.Compared with Fast independent component analysis and Single Channel algorithms,chrominance-based method and plane-orthogonal-to-skin algorithms have a more vital anti-interference ability and higher robustness.The performances of the five-organs fusion area and the full-face area were better than that of single sub-regions,and the fewer motion artifacts and better lighting can improve the precision of pulse rate estimation.

Faster Region Convolutional Neural Network(FRCNN)Based Facial Emotion Recognition

Facial emotion recognition(FER)has become a focal point of research due to its widespread applications,ranging from human-computer interaction to affective computing.While traditional FER techniques have relied on handcrafted features and classification models trained on image or video datasets,recent strides in artificial intelligence and deep learning(DL)have ushered in more sophisticated approaches.The research aims to develop a FER system using a Faster Region Convolutional Neural Network(FRCNN)and design a specialized FRCNN architecture tailored for facial emotion recognition,leveraging its ability to capture spatial hierarchies within localized regions of facial features.The proposed work enhances the accuracy and efficiency of facial emotion recognition.The proposed work comprises twomajor key components:Inception V3-based feature extraction and FRCNN-based emotion categorization.Extensive experimentation on Kaggle datasets validates the effectiveness of the proposed strategy,showcasing the FRCNN approach’s resilience and accuracy in identifying and categorizing facial expressions.The model’s overall performance metrics are compelling,with an accuracy of 98.4%,precision of 97.2%,and recall of 96.31%.This work introduces a perceptive deep learning-based FER method,contributing to the evolving landscape of emotion recognition technologies.The high accuracy and resilience demonstrated by the FRCNN approach underscore its potential for real-world applications.This research advances the field of FER and presents a compelling case for the practicality and efficacy of deep learning models in automating the understanding of facial emotions.

Global-local combined features to detect pain intensity from facial expression images with attention mechanism

The estimation of pain intensity is critical for medical diagnosis and treatment of patients.With the development of image monitoring technology and artificial intelligence,automatic pain assessment based on facial expression and behavioral analysis shows a potential value in clinical applications.This paper reports a framework of convolutional neural network with global and local attention mechanism(GLA-CNN)for the effective detection of pain intensity at four-level thresholds using facial expression images.GLA-CNN includes two modules,namely global attention network(GANet)and local attention network(LANet).LANet is responsible for extracting representative local patch features of faces,while GANet extracts whole facial features to compensate for the ignored correlative features between patches.In the end,the global correlational and local subtle features are fused for the final estimation of pain intensity.Experiments under the UNBC-McMaster Shoulder Pain database demonstrate that GLA-CNN outperforms other state-of-the-art methods.Additionally,a visualization analysis is conducted to present the feature map of GLA-CNN,intuitively showing that it can extract not only local pain features but also global correlative facial ones.Our study demonstrates that pain assessment based on facial expression is a non-invasive and feasible method,and can be employed as an auxiliary pain assessment tool in clinical practice.

Re-Distributing Facial Features for Engagement Prediction with ModernTCN

Automatically detecting learners’engagement levels helps to develop more effective online teaching and assessment programs,allowing teachers to provide timely feedback and make personalized adjustments based on students’needs to enhance teaching effectiveness.Traditional approaches mainly rely on single-frame multimodal facial spatial information,neglecting temporal emotional and behavioural features,with accuracy affected by significant pose variations.Additionally,convolutional padding can erode feature maps,affecting feature extraction’s representational capacity.To address these issues,we propose a hybrid neural network architecture,the redistributing facial features and temporal convolutional network(RefEIP).This network consists of three key components:first,utilizing the spatial attention mechanism large kernel attention(LKA)to automatically capture local patches and mitigate the effects of pose variations;second,employing the feature organization and weight distribution(FOWD)module to redistribute feature weights and eliminate the impact of white features and enhancing representation in facial feature maps.Finally,we analyse the temporal changes in video frames through the modern temporal convolutional network(ModernTCN)module to detect engagement levels.We constructed a near-infrared engagement video dataset(NEVD)to better validate the efficiency of the RefEIP network.Through extensive experiments and in-depth studies,we evaluated these methods on the NEVD and the Database for Affect in Situations of Elicitation(DAiSEE),achieving an accuracy of 90.8%on NEVD and 61.2%on DAiSEE in the fourclass classification task,indicating significant advantages in addressing engagement video analysis problems.

Facial Image-Based Autism Detection:A Comparative Study of Deep Neural Network Classifiers

Autism Spectrum Disorder(ASD)is a neurodevelopmental condition characterized by significant challenges in social interaction,communication,and repetitive behaviors.Timely and precise ASD detection is crucial,particularly in regions with limited diagnostic resources like Pakistan.This study aims to conduct an extensive comparative analysis of various machine learning classifiers for ASD detection using facial images to identify an accurate and cost-effective solution tailored to the local context.The research involves experimentation with VGG16 and MobileNet models,exploring different batch sizes,optimizers,and learning rate schedulers.In addition,the“Orange”machine learning tool is employed to evaluate classifier performance and automated image processing capabilities are utilized within the tool.The findings unequivocally establish VGG16 as the most effective classifier with a 5-fold cross-validation approach.Specifically,VGG16,with a batch size of 2 and the Adam optimizer,trained for 100 epochs,achieves a remarkable validation accuracy of 99% and a testing accuracy of 87%.Furthermore,the model achieves an F1 score of 88%,precision of 85%,and recall of 90% on test images.To validate the practical applicability of the VGG16 model with 5-fold cross-validation,the study conducts further testing on a dataset sourced fromautism centers in Pakistan,resulting in an accuracy rate of 85%.This reaffirms the model’s suitability for real-world ASD detection.This research offers valuable insights into classifier performance,emphasizing the potential of machine learning to deliver precise and accessible ASD diagnoses via facial image analysis.

Recurrent herpes infection showing a new facial phenotype:A case report

BACKGROUND Facial herpes is a common form of the herpes simplex virus-1 infection and usually presents as vesicles near the mouth,nose,and periocular sites.In contrast,we observed a new facial symptom of herpes on the entire face without vesicles.CASE SUMMARY A 33-year-old woman with a history of varicella infection and shingles since an early age presented with sarcoidosis of the entire face and neuralgia without oral lesions.The patient was prescribed antiviral treatment with valacyclovir and acyclovir cream.One day after drug administration,facial skin lesions and neurological pain improved.Herpes simplex without oral blisters can easily be misdiagnosed as pimples upon visual examination in an outpatient clinic.CONCLUSION As acute herpes simplex is accompanied by neuralgia,prompt diagnosis and prescription are necessary,considering the pathological history and health conditions.

Evaluation of Ear and Facial Indices of Ibibio and Efik Female Children of Akwa Ibom and Cross River States, Nigeria

Background: The ear and face are indispensable and distinctive features for hearing and identification. Objectives: This study was designed to generate anthropometric data of the ear and facial indices of females of Efik and Ibibio children in Cross River and Akwa Ibom States, show morphological and aesthetic differences and ethnicity. Methods: A total of 600 female children (300 Efiks and 300 Ibibios) aged 2 to 10 years that met the inclusion criteria were chosen from selected primary schools in Calabar Municipality, Calabar South of Cross River State and from Uyo, Itu of Akwa Ibom State, Nigeria. Standardized measurements of face length, face width, ear length, and ear width were taken with a spreading caliper;the facial (proscopic) and ear (auricular) indices were determined. Results: Efik subjects presented a mean face length of 8.36 ± 0.06 cm, face width of 11.04 ± 0.04 cm, ear length of 4.92 ± 0.02 cm, and ear width of 3.06 ± 0.01 cm. Ibibio subjects had mean values for face length, face width, ear length, and ear width as 8.17 ± 0.05 cm, 10.75 ± 0.05 cm, 4.77 ± 0.03 cm, and 2.94 ± 0.02 cm respectively. The mean facial index and ear index for Efik subjects were 75.68 ± 0.31 and 62.16 ± 0.27 respectively;while the mean facial and ear indices for Ibibio subjects were 74.79 ± 0.36 and 61.80 ± 0.34 respectively. Statistical analysis demonstrated significant differences in face length, ear length, ear width and facial index, with the Efik subjects having higher values than Ibibio subjects (p Conclusion: The results showed hypereuryproscopic face as the prevalent face type among females of both ethnic groups, therefore can be of importance in sex, ethnic, and racial differentiation, and in clinical practice, aesthetics and forensic medicine.

A Robust Method of Bipolar Mental Illness Detection from Facial Micro Expressions Using Machine Learning Methods

Bipolar disorder is a serious mental condition that may be caused by any kind of stress or emotional upset experienced by the patient.It affects a large percentage of people globally,who fluctuate between depression and mania,or vice versa.A pleasant or unpleasant mood is more than a reflection of a state of mind.Normally,it is a difficult task to analyze through physical examination due to a large patient-psychiatrist ratio,so automated procedures are the best options to diagnose and verify the severity of bipolar.In this research work,facial microexpressions have been used for bipolar detection using the proposed Convolutional Neural Network(CNN)-based model.Facial Action Coding System(FACS)is used to extract micro-expressions called Action Units(AUs)connected with sad,happy,and angry emotions.Experiments have been conducted on a dataset collected from Bahawal Victoria Hospital,Bahawalpur,Pakistan,Using the Patient Health Questionnaire-15(PHQ-15)to infer a patient’s mental state.The experimental results showed a validation accuracy of 98.99%for the proposed CNN modelwhile classification through extracted featuresUsing SupportVectorMachines(SVM),K-NearestNeighbour(KNN),and Decision Tree(DT)obtained 99.9%,98.7%,and 98.9%accuracy,respectively.Overall,the outcomes demonstrated the stated method’s superiority over the current best practices.

Sensory Evaluation of Freeze-Dried Facial Mask and Analysis of User Profile

To thoroughly understand market opportunity of freeze-dried facial mask and deeply get insight of consumers’usage behavior and needs,evaluate sensory feelings of 10 screened commercial freeze-dried facial mask products,group test products according to the differences of sensory attributions via Principal Component Analysis(PCA)and Agglomerative Hierarchical Clustering(AHC),pick up the representative products.Freeze-dried facial mask users evaluate satisfaction degree of picked up products and participate survey of usage behavior/cognition.Analyze consumer data by AHC to get consumer segmentations and their profile.The test results show that,sensory data and consumer data,which is from consumers test of screened representative products by performing PCA and AHC on sensory data,can be verified mutually.It is helpful to understand the needs of consumer segmentations and reason to buy by combining sensory data and consumer test.

Epidemiological and Computed Tomography Aspects of Facial Trauma at Kira Hospital in Bujumbura

Background: Maxillofacial trauma affects young adults more. The injury assessment is difficult to establish in low-income countries because of the imaging means, particularly the scanner, which is poorly available and less financially accessible. The aim of this study is to describe the epidemiological profile and the various tomodensitometric aspects of traumatic lesions of the face in patients received in the Radiology department of Kira Hospital. Patients and methods: This is a descriptive retrospective study involving 104 patients of all ages over a period of 2 years from December 2018 to November 2019 in the medical imaging department of KIRA HOSPITAL. We included in our study any patient having undergone a CT scan of the head and presenting at least one lesion of the facial mass, whether associated with other cranioencephalic lesions. Results: Among the 384 patients received for head trauma, 104 patients (27.1% of cases) presented facial damage. The average age of our patients was 32.02 years with extremes of 8 months and 79 years. In our study, 87 of the patients (83.6%) were male. The road accident was the circumstance in which facial trauma occurred in 79 patients (76% of cases). These injuries were accompanied by at least one bone fracture in 97 patients (93.3%). Patients with fractures of more than 3 facial bones accounted for 40.2% of cases and those with fractures of 2 to 3 bones accounted for 44.6% of cases. The midface was the site of the fracture in 85 patients (87.6% of cases). Orbital wall fractures were noted in 57 patients (58.8% of cases) and the jawbone was the site of a fracture in 50 patients (51.5% of cases). In the vault, the fractures involved the extra-facial frontal bone (36.1% of cases) and temporal bone (18.6% of cases). Cerebral contusion was noted in 41.2% of patients and pneumoencephaly in 15.5% of patients. Extradural hematoma was present in 16 patients and subdural hematoma affected 13 patients. Conclusion: Computed tomography is a diagnostic tool of choice in facial trauma patients. Most of these young patients present with multiple fractures localizing to the mid-level of the face with concomitant involvement of the brain.

Peripheral Facial Paralysis in People Living with Human Immunodeficiency Virus (HIV)

Introduction: Peripheral facial palsy (PFP) is a frequent reason for ENT consultations. It is a common complication of human immunodeficiency virus (HIV) infection. The aim of this study was to describe the diagnostic and therapeutic aspects and to establish the correlation between PFP and HIV in our context. Patients and Method: This was a retrospective descriptive study conducted in the ENT and CFS department of the HIAOBO, covering the medical records of patients hospitalized for taking a PFP on HIV terrain from January 1, 2016 to December 31, 2020. Results: The study involved 17 patients, 10 men (59%) and 7 women (41%), a sex ratio of 1.4. The average age was 39 years with the extremes of 11 and 69 years. Shopkeepers reported 9 cases (53%). The reason for consultation was facial asymmetry in 11 cases (100%). The delay in consultation during the first week was 82.4%. Clinical signs were unilateral facial asymmetry, the opening of the palpebral fissure and lacrimation. All patients received medical treatment for PFP and HIV. Evolution was favorable, with complete recovery and no sequelae in 82.4% of cases. Surgery was performed in one case. Conclusion: PFPs are common in HIV infection. Diagnosis is clinical and management is multidisciplinary. Progression depends on the length of time taken to treat the disease.

Facial expression recognition based on fuzzy-LDA/CCA

A novel fuzzy linear discriminant analysis method by the canonical correlation analysis （fuzzy-LDA/CCA）is presented and applied to the facial expression recognition. The fuzzy method is used to evaluate the degree of the class membership to which each training sample belongs. CCA is then used to establish the relationship between each facial image and the corresponding class membership vector, and the class membership vector of a test image is estimated using this relationship. Moreover, the fuzzy-LDA/CCA method is also generalized to deal with nonlinear discriminant analysis problems via kernel method. The performance of the proposed method is demonstrated using real data.

PROF.ZHANG ANLI'S EXPERIENCE OF TREATING STUBBORN CASES OF FACIAL PALSY

Based on her own experience of many years＇ clinical practice, Prof. Zhang Anli summarized and created the method in which the shallow insertion at the upper eyelid, acupuncture at the Back-shu points and balanced selection of points together with the modified Setting Mountain on Fire and application of Fire Needle were applied for the treatment of stubborn cases of facial palsy. The therapeutic effect is good.

Facial pain induces the alteration of transient receptor potential vanilloid receptor 1 expression in rat trigeminal ganglion

Objective To investigate the involvement of transient receptor potential vanilloid receptor 1 （TRPV1） in the facial inflammatory pain in relation to thermal hyperalgesia and cold pain sensation. Methods Facial inflammatory pain model was developed by subcutaneous injection of turpentine oil （TO） into rat facial area. Head withdrawal thermal latency （HWTL） and head withdrawal cold latency （HWCL） were measured once a day for 21 d after TO treatment using thermal and cold measurement apparatus. The immunohistochemical staining, cell-size frequency analysis and the survey of average optical density （OD） value were used to observe the changes of TRPV1 expression in the neurons of the trigeminal ganglion （TG）, peripheral nerve fibers in the vibrissal pad, and central projection processes in the trigeminal sensory nuclei caudalis （Vc） on day 3, 5, 7, 14, and 21 after TO injection. Results HWTL and HWCL decreased significantly from day 1 to day 14 after TO injection with the lowest value on day 5 and day 3, respectively, and both recovered on day 21. The number of TRPV1-labeled neurons increased remarkably from day 1 to day 14 with a peak on day 7, and returned back to the normal level on day 21. In control rats, only small and medium-sized TG neurons were immunoreactive （IR） to TRPV1, and the TRPV1-IR terminals were abundant in both the vibrissal pad and the Vc. Within 2 weeks of inflammation, the expression of TRPV1 in small and medium-sized TG neurons increased obviously. Also the TRPV1 stained terminals and fibers appeared more frequent and denser in both the vibrissal pad skin and throughout laminae Ⅰ and the outer zone of laminae Ⅱ （Ⅱo） of Vc. Conclusion Facial inflammatory pain could induce hyperalgesia to noxious heat and cold stimuli, and result in increase of the numbers of TRPV1 positive TG neurons and the peripheral and central terminals of TG. These results suggest that the phenotypic changes of TRPV1 expression in small and medium-sized TG neurons and terminals might play an important role in the development and maintenance of TO-induced inflammatory thermal hyperalgesia and cold pain sensation.

Bilateral Peripheral Facial Paralysis Combined with HIV Meningitis During Acute HIV-1 Infection: A Case Report

Here we reported a Chinese case of bilateral peripheral facial paralysis(PFP) in human immunodeficiency virusc(HIV) infected population. A 38-year-old homosexual male patient was referred to our hospital for bilateral facial paralysis. 21 days prior to admission he had developed high fever, chills, headache, fatigue, general malaise, nausea and vomiting. Neurological examination revealed bilateral ptosis of lower lip and cheeks, as well as failure of bilateral eyes closure. Analysis of cerebrospinal fluid(CSF) revealed pleocytosis, a marked rise of micro total protein and a marked rise of intrathecal lgG synthesis. The result of HIV-1 serology was positive by ELISA and that was confirmed by western blot. His CD4^+ cell count was 180 cells/mm^3. HIV-1 viral load in CSF was almost 10 times higher than that in plasma. The patient's condition improved steadily and experienced complete resolution of bilateral PFP after 2 months.

Electroacupuncture promotes peripheral nerve regeneration after facial nerve crush injury and upregulates the expression of glial cell-derived neurotrophic factor

The efficacy of electroacupuncture in the treatment of peripheral facial paralysis is known, but the specific mechanism has not been clarified. Glial cell-derived neurotrophic factor(GDNF) has been shown to protect neurons by binding to N-cadherin. Our previous results have shown that electroacupuncture could increase the expression of N-cadherin mRNA in facial neurons and promote facial nerve regeneration. In this study, the potential mechanisms by which electroacupuncture promotes nerve regeneration were elucidated through assessing the effects of electroacupuncture on GDNF and N-cadherin expression in facial motoneurons of rabbits with peripheral facial nerve crush injury. New Zealand rabbits were randomly divided into a normal group(normal control, n = 21), injury group(n = 45) and electroacupuncture group(n = 45). Model rabbits underwent facial nerve crush injury only. Rabbits in the electroacupuncture group received facial nerve injury, and then underwent electroacupuncture at Yifeng(TE17), Jiache(ST6), Sibai(ST2), Dicang(ST4), Yangbai(GB14), Quanliao(SI18), and Hegu(LI4; only acupuncture, no electrical stimulation). The results showed that in behavioral assessments, the total scores of blink reflex, vibrissae movement, and position of apex nasi, were markedly lower in the EA group than those in the injury group. Hematoxylin-eosin staining of the right buccinator muscle of each group showed that the cross-sectional area of buccinator was larger in the electroacupuncture group than in the injury group on days 1, 14 and 21 post-surgery. Toluidine blue staining of the right facial nerve tissue of each group revealed that on day 14 post-surgery, there was less axonal demyelination and fewer inflammatory cells in the electroacupuncture group compared with the injury group. Quantitative real time-polymerase chain reaction showed that compared with the injury group, N-cadherin mRNA levels on days 4, 7, 14 and 21 and GDNF mRNA levels on days 4, 7 and 14 were significantly higher in the electroacupuncture group. Western blot assay displayed that compared with the injury group, the expression of GDNF protein levels on days 7, 14 and 21 were significantly upregulated in the electroacupuncture group. The histology with hematoxylin-eosin staining and Nissl staining of brainstem tissues containing facial neurons in the middle and lower part of the pons exhibited that on day 7 post-surgery, there were significantly fewer apoptotic neurons in the electroacupuncture group than in the injury group. By day 21, there was no significantly difference in the number of neurons between the electroacupuncture and normal groups. Taken together, these results have confirmed that electroacupuncture promotes regeneration of peripheral facial nerve injury in rabbits, inhibits neuronal apoptosis, and reduces peripheral inflammatory response, resulting in the recovery of facial muscle function. This is achieved by up-regulating the expression of GDNF and N-cadherin in central facial neurons.

Patterns of cortical reorganization in facial synkinesis:a task functional magnetic resonance imaging study

Facial synkinesis,a sequela of peripheral facial nerve palsy,is characterized by simultaneous involuntary facial movement during a voluntary desired one.Maladaptive cortical plasticity might be involved in the dysfunction of facial muscles.This cohort study investigated the cortical functional alterations in patients with unilateral facial synkinesis,using the task functional magnetic resonance imaging.Facial motor tasks,including blinking and smiling,were performed by 16 patients（aged 30.6 ± 4.5 years,14 females/2 males） and 24 age-and sex-matched healthy controls（aged 29.1 ± 4.2 years,19 females/5 males）.Results demonstrated that activation in the cortico-facial motor representation area was lower during tasks in patients with facial synkinesis compared with healthy controls.Facial movements on either side performed by patients caused more intensive activation of the supplementary motor area on the contralateral side of the affected face,than those on the unaffected side.Our results revealed that there was cortical reorganization in the primary sensorimotor area and the supplementary motor area.This study was registered in Chinese Clinical Trial Registry（registration number： Chi CTR1800014630）.

Identity-aware convolutional neural networks for facial expression recognition

Facial expression recognition is a hot topic in computer vision, but it remains challenging due to the feature inconsistency caused by person-specific 'characteristics of facial expressions. To address such a challenge, and inspired by the recent success of deep identity network (DeepID-Net) for face identification, this paper proposes a novel deep learning based framework for recognising human expressions with facial images. Compared to the existing deep learning methods, our proposed framework, which is based on multi-scale global images and local facial patches, can significantly achieve a better performance on facial expression recognition. Finally, we verify the effectiveness of our proposed framework through experiments on the public benchmarking datasets JAFFE and extended Cohn-Kanade (CK+).

Characterization of neurovascular compression in facial neuralgia patients by 3D high-resolution MRI and image fusion technique

Objective:To describe the anatomical characteristics and patterns of neurovascular compression (NVC) in patients suffering trigeminal neuralgia(TN) by 3D high-resolution magnetic resonance imaging(MRI) method and image fusion technique.Methods:The anatomic structure of trigeminal nerve,brain stem and blood vessel was observed in 100 consecutive TN patients by 3D high resolution MRI(3D SPGR,contrast-enhanced T1 3D MP-RAGE and T2/T1 3D FIESTA). The 3D image sources were fused and visualized using 3D DOCTOR software.Results:One or several NVC sites,which usually appeared 0-9.8 mm away from brain stem,were found on the symptomatic side in 93%of the TN cases.Superior cerebellar artery was involved in 76%(71/93) of these cases.The other vessels including antero-inferior cerebellar artery,vertebral artery, basilar artery and veins also contributed to the occurrence of NVC.The NVC sites were found to be located in the proximal segment in 42%of these cases(39/93) and in the distal segment in 45% (42/93).Nerve dislocation or distortion was observed in 32%(30/93).Conclusions:Various 3D high resolution MRI methods combined with the image fusion technique could provide pathologic anatomic information for the diagnosis and treatment of TN.

REVIEW ON ACUPUNCTURE TREATMENT OF PERIPHERAL FACIAL PARALYSIS DURING THE PAST DECADE

Facial paralysis or Bell’s palsy is acommon disease that frequently attacksadults aged 20--40 years, especially women.According to traditional Chinese medicine,the pathogenesis is believed to be due to in-vasion of the channels and collaterals