Abstract We developed an automatic method for staging periodontitis on dental panoramic radiographs using the deep learning hybrid method. A novel framework was proposed to automatically detect and classify periodontal bone loss of each individual tooth. The is a architecture detection conventional CAD processing classification. Deep used radiographic level (or CEJ level) as simple structure whole jaw radiographs. Next, percentage rate analysis combined tooth long-axis with levels. Using...

Abstract The purpose of this study was to directly and quantitatively measure BMD from Cone-beam CT (CBCT) images by enhancing the linearity uniformity bone intensities based on a hybrid deep-learning model (QCBCT-NET) combining generative adversarial network (Cycle-GAN) U-Net, compare enhanced QCBCT-NET with those Cycle-GAN U-Net. We used two phantoms human skulls encased in acrylic, one for training validation datasets, other test dataset. proposed consisting residual blocks multi-channel...

Abstract Objective 4‐n‐butylresorcinol is a competitive inhibitor of tyrosinase and has been used as an antimelanogenic agent. However, its inhibition mechanism in intact cells not fully understood. To elucidate the cellular mechanism, we compared vitro vivo inhibitory effects on activity. Methods B16F10 melanoma were cultured media containing α ‐ MSH presence or absence 4‐n‐butylresorcinol. Tyrosinase mRNA levels, protein levels activity by real‐time PCR , immunostaining combined with...

The purpose of this study was to develop a complete digital workflow for planning, simulation, and evaluation orthognathic surgery based on 3D natural head position reproduction, cloud-based collaboration platform, landmark-based evaluation. We included 24 patients who underwent bimaxillary surgery. Surgeons engineers could share the massive image data immediately conveniently collaborate closely in surgical planning simulation using platform. splint be optimized specific patient before or...

The goal of this work is to provide an abstraction ideal sound environments a new emerging class Mobile Multi-speaker Audio (MMA) applications. Typically, it challenging for MMA applications implement advanced features (e.g., surround sound) accurately in mobile environments, especially due unknown, irregular loudspeaker configurations. Towards illusion that run over specific configurations (i.e., speaker type, layout), proposes AMAC, Adaptive Coordination system senses the acoustic...

Aspect-level news browsing provides readers with a classified view of articles different viewpoints. It facilitates active interactions which easily discover and compare diverse existing biased views over event. As such, it effectively helps understand the event from plural viewpoints formulate their own, more balanced free specific views. Realizing aspect-level raises important challenges, mainly due to lack semantic knowledge abstract classify intended salient aspects articles. We first...

The purpose of this study was to develop a quantitative AR-assisted free-hand orthognathic surgery method using electromagnetic (EM) tracking and skin-attached dynamic reference. authors proposed novel, simplified, convenient workflow for augmented reality (AR)-assisted based on optical marker-less tracking, comfortable display, non-invasive, reference frame. 2 registrations between the physical (EM tracking) CT image spaces AR camera spaces, essential processes in surgery, were...

Increasing demand for on-device Automatic Speech Recognition (ASR) systems has resulted in renewed interests developing automatic model compression techniques. Past research have shown that AutoML-based Low Rank Factorization (LRF) technique, when applied to an end-to-end Encoder-Attention-Decoder style ASR model, can achieve a speedup of up 3.7x, outperforming laborious manual rank-selection approaches. However, we show current search techniques only work certain level, beyond which they...

As location-aware mobile devices such as smartphones have now become prevalent, people are able to easily record their trajectories in daily lives. Such personal a very promising means share life experiences, since important contextual information significant locations and activities can be extracted from the raw trajectories. In this paper, we propose MetroScope, trajectory-based real-time on-the-go experience sharing system metropolitan city. MetroScope allows experiences through...

Abstract Panoramic radiography is the most commonly used equipment in dental field, but there no comprehensive standard about how to evaluate spatial resolution of panoramic radiography. In this study, panorama phantoms were developed for evaluation horizontal and vertical reflecting unique characteristics Four staircase shape designed obtain images lead line pairs a 52 mm width. with placed at an oblique angle also designed. A phantom stand was made. Three machines evaluated twice by two...

This paper developed a new method to easily record and automatically reproduce the 3D natural head position (NHP) of patients using portable scanner based on immediate calibration. We first optically scanned patient’s face scanner, model was aligned with global horizon an calibration procedure plate. The patient NHP Computed Tomography(CT) reproduced by performing registration between CT in modified coherent point drift (CPD) algorithm. In phantom experiment, we evaluated method’s accuracy...

A variety of advantages from sounds such as measurement and accessibility introduces a new opportunity for mobile applications to offer broad types interesting, valuable functionalities, supporting richer user experience. However, in spite the growing interests on sound applications, few or no works have been done focusing managing an audio device effectively. More specifically, their low level real-time capability resources makes it challenging satisfy tight timing requirements e.g., high...

The paradigm of the Internet Things predicts that not only typical computing devices but also mundane objects will soon cooperate to achieve certain goals and communicate with users. In this paper we present a new way designing multi-device interaction accommodate diverse types relations among them. We first held design workshop wherein participants created ideas by using device cards. then analyzed in terms roles devices, user/environment, data type flow direction. Based on findings,...

Users face various usability issues when interacting with mobile apps. While collecting quality in-situ feedback is extremely valuable for the developers UX improvement, our understanding of user reporting behavior and interface design very limited. In this paper, we conducted a study to solicit from five apps using think-aloud protocol video recordings. Using collected data, investigated attributes related behavior. We identified four key behavior, i.e., issue topics, origin issues, types...

The blood glucose prediction algorithms were widely studied for the patients with type 1 diabetes. We analyzed data of 16 2 diabetic underwent continuous monitoring (CGM) through deep-learning models. Feed Forward Neural Network (FFNN) and Long Short-Term Memory (LSTM) used to model predict glucose. Blood levels after 5, 15, 30, 45 min predicted each patient. To evaluate predictive power model, Root Mean Square Error (RMSE) was calculated by comparing actual value RMSE FFNN lower than that...

This paper aims to provide a selective survey about knowledge distillation(KD) framework for researchers and practitioners take advantage of it developing new optimized models in the deep neural network field. To this end, we give brief overview distillation some related works including learning using privileged information(LUPI) generalized distillation(GD). Even though based on teacher-student architecture was initially devised as model compression technique, has found versatile...

Summary form only given. The aim of this study was to develop a robot-assisted system for orthognathic surgery and evaluate robot movement. registration procedure performed in coordinates among robot, an optical tracking (OTS) CT image with fiducials. According the surgical plan coordinate, arm moved corresponding position. results were below 0.15mm translational movement 0.41mm rotational

The principal objective of this study was to develop a scanning technique that helps patients reduce their anxiety and relax physical tension in the MRI system. targeted 10 healthy persons with no medical history disorder past current clinically diagnosed disorder, as well who were currently experiencing an during scanning. focusing board assembly self-manufactured conduct clinical experiment via scans. As method confirm efficacy experiment, bio meter used measure brainwaves from targets...