Intraoral X-ray Imaging Research Articles

CVApool: using null-space of CNN weights for the tooth disease classification

In light of current developments in dental care, dental professionals have increasingly used deep learning methods to get precise diagnoses of oral problems. Using intraoral X-rays in dental radiography is imperative in many dental interventions. Integrating deep learning techniques with a unique collection of intraoral X-ray images has been undertaken to enhance the accuracy of dental disease detection. In this study, we propose an alternative pooling layer, namely the Common Vector Approach Pooling technique, to address the constraints associated with average pooling in deep learning methods. The experiments are conducted on a large dataset, involving twenty different dental conditions, divided into seven categories. Our proposed approach achieved a high accuracy rate of 86.4% in identifying dental problems across the seven oral categories.

Influence of the Tube Angle on the Measurement Accuracy of Peri-Implant Bone Defects in Rectangular Intraoral X-ray Imaging.

Intraoral radiography in the right-angle technique is the standard procedure to examine the peri-implant bone level in implant follow-up and implant-related studies. For the implementation of the right-angle or parallel technique, mostly ready-made image receptor holders are used. The aim of this experimental study is to analyze changes in the measurement of standardized peri-implant defects caused by a deviation in the position of the image receptor. Eleven Xive® implants (Dentsply Sirona, Bensheim, Germany) were placed in bovine bone, and peri-implant defects of varying depths were created. The preparations were fixed in a specially made test stand, and intraoral radiographs were taken using the right-angle technique with standard film holders at various horizontal and vertical projection angles. Defect measurement was carried out with the imaging software Sidexis 4 V 4.3 (Dentsply Sirona, Bensheim, Germany). With increasing angular deviation, larger deviations between the measured and the real extent of the defect occurred. Vertical tilting caused significant distortion, while horizontal rotation showed less effect. Intraoral radiography only provides a valid representation of the peri-implant bone level for follow-up or as a tool in implant-related studies if a reproducible projection direction is assured.

A Survey of Dental Caries Segmentation and Detection Techniques.

Dental caries detection, in the past, has been a challenging task given the amount of information got from various radiographic images. Several methods have been introduced to improve the quality of images for faster caries detection. Deep learning has become the methodology of choice when it comes to analysis of medical images. This survey gives an in-depth look into the use of deep learning for object detection, segmentation, and classification. It further looks into literature on segmentation and detection methods of dental images through deep learning. From the literature studied, we found out that methods were grouped according to the type of dental caries (proximal, enamel), type of X-ray images used (extraoral, intraoral), and segmentation method (threshold-based, cluster-based, boundary-based, and region-based). From the works reviewed, the main focus has been found to be on threshold-based segmentation methods. Most of the reviewed papers have preferred the use of intraoral X-ray images over extraoral X-ray images to perform segmentation on dental images of already isolated parts of the teeth. This paper presents an in-depth analysis of recent research in deep learning for dental caries segmentation and detection. It involves discussing the methods and algorithms used in segmenting and detecting dental caries. It also discusses various existing models used and how they compare with each other in terms of system performance and evaluation. We also discuss the limitations of these methods, as well as future perspectives on how to improve their performance.

Vertical root fracture diagnosis in endodontically treated teeth

Annotation. Timely diagnosis of vertical root fractures is a big challenge to dentists, because there are no specific clinical signs, symptoms and X-ray features. The aim of the research is to study clinical cases of vertical root fractures in endodontically treated teeth and to carry out analysis the methods of their diagnosis, clinical features and radiological characteristics. In patients aged 40-65 years, 34 clinical cases of vertical root fracture are considered. The main and additional methods of patients’ examination are analyzed. The statistical calculation of the obtained results is performed by the method of variation statistics with the determination of the Student’s t-test using computer programs Microsoft® Excel 2017 and “Statistica 6.1”. Clinical symptoms and abnormal findings during examination were recorded but not subjected to statistical analysis. Clinical examination revealed periodontal swelling or abscess in 6 patients, isolated narrow and deep periodontal pocket in 24 patients, low localized fistula in 9 patients. Radiological results: only in 6 clinical cases a vertical fracture line was observed on intraoral X-ray images; in 30 cases the destruction of bone tissue in the periapical area, mainly in the halo-type radiolucency was determined; in 4 cases no pathological changes in periapical bone tissue were revealed. Vertical root fracture was confirmed by muco-periosteal flap detachment in all cases. It has been established that a number of signs, namely: isolated deep narrow periodontal pocket; coronally located fistula; characteristic bone destruction (J-shaped or “halo” radiolucency) on the radiograph indicate the likelihood of vertical root fracture. Signs of fenestration of the bone adjacent to the fractured root and visualization of the fracture line after mucoperiosteal flap surgery are pathognomonic.

Classification of Intraoral X-ray Images Using Artificial Intelligence

Classification of Intraoral X-ray Images Using Artificial Intelligence

Study of the Multi-Directional Intraoral X-Ray System for Clinical Diagnosis of Dental Caries

For minimizing patient inconvenience caused by intraoral sensor discomfort and unnecessary radiational exposure of diagnosis of dental caries, we developed a new system that insert the x-ray source tube (diameter: 3 mm, tube voltage: ~ 55 kV, tube current: ~ 300 uA) inside of the oral cavity and placed the the radiation acquisition device outside. The purpose of this study was to investigate the optimal acquisition conditions for the diagnosis of dental caries and to evaluate the image performance of the new intraoral x-ray imaging system. The 33 model sets were made with 3 teeth were randomly selected from 99 extracted human teeth, total of 5 conventional radiographic sets as control groups and 1 experimental set were prepared after cavity forming the 4 sites of artificial proximal caries(grade 1 to 4). Total 198 images were obtained and those were read by seven dentists of the grades and the correct answer rates were calculated and analyzed by one - way ANOVA. The new x-ray images were shown that there is no significant difference in reading for the diagnosis of proximal caries. Therefore the new x-ray system might be considerable for using in the clinical dental caries diagnosis, especially initial caries stage.

Control of the Low-energy X-rays by Using MCNP5 and Numerical Analysis for a New Concept Intra-oral X-ray Imaging System

An X-ray control algorithm to modulate the X-ray intensity distribution over the FOV (field of view) has been developed by using numerical analysis and MCNP5, a particle transport simulation code on the basis of the Monte Carlo method. X-rays, which are widely used in medical diagnostic imaging, should be controlled in order to maximize the performance of the X-ray imaging system. However, transporting X-rays, like a liquid or a gas is conveyed through a physical form such as pipes, is not possible. In the present study, an X-ray control algorithm and technique to uniformize the Xray intensity projected on the image sensor were developed using a flattening filter and a collimator in order to alleviate the anisotropy of the distribution of X-rays due to intrinsic features of the X-ray generator. The proposed method, which is combined with MCNP5 modeling and numerical analysis, aimed to optimize a flattening filter and a collimator for a uniform distribution of X-rays. Their size and shape were estimated from the method. The simulation and the experimental results both showed that the method yielded an intensity distribution over an X-ray field of 6×4 cm2 at SID (source to image-receptor distance) of 5 cm with a uniformity of more than 90% when the flattening filter and the collimator were mounted on the system. The proposed algorithm and technique are not only confined to flattening filter development but can also be applied for other X-ray related research and development efforts.

Clinical Evaluation of Bioabsorbable Mesh for Secondary Bone Grafts in the Alveolar Cleft

Purpose: This study investigates whether a poly-L-lactic acid (PLLA)-polyglycolic acid (PGA) mesh-type bone joining material (PLLA-PGA mesh), a bioabsorbable osteosynthetic material, can replace titanium mesh for secondary bone grafts in the alveolar cleft. Methods: The subjects were 6 patients who underwent secondary bone grafting in the alveolar cleft using the PLLA-PGA mesh and 6 patients treated with titanium mesh. Intraoral X-ray images and dental small-field CT (3DX) were taken over a period of 6 months postoperatively. Analysis of Covariance (ANCOVA) was performed for two objective variables obtained from CT images namely the vertical height of the bony bridge and the labial-lingual thickness. A generalized linear model was also used to analyze a discrete objective variable, i.e., the score for the bony bridge height obtained from the occlusal X-ray image. In these analyses, the five explanatory variables used were mesh type, sex, age, grafted bone mass, and width of the alveolar cleft measured on the CT image. Results: No significant factors were observed among the 5 explanatory variables with regard to influence on bony bridge thickness. Mesh type was the only significant factor influencing bony bridge height, and the height was greater when using PLLA-PGA mesh than when using titanium mesh. However, there were no significant factors influencing alveolar crest height on occlusal X-ray images. Conclusions: There were no marked differences in bony bridge morphology in the alveolar cleft area between the PLLA-PGA and titanium mesh groups.

Using fractal dimension to evaluate alveolar bone defects treated with various bone substitute materials

Abstract

The Effects on Absorbed Dose Distribution in Intraoral X-ray Imaging When Using Tube Voltages of 60 and 70 kV for Bitewing Imaging

ABSTRACTObjectivesEfforts are made in radiographic examinations to obtain the best image quality with the lowest possible absorbed dose to the patient. In dental radiography, the absorbed dose to patients is very low, but exposures are relatively frequent. It has been suggested that frequent low-dose exposures can pose a risk for development of future cancer. It has previously been reported that there was no significant difference in the diagnostic accuracy of approximal carious lesions in radiographs obtained using tube voltages of 60 and 70 kV. The aim of this study was, therefore, to evaluate the patient dose resulting from exposures at these tube voltages to obtain intraoral bitewing radiographs.Material and MethodsThe absorbed dose distributions resulting from two bitewing exposures were measured at tube voltages of 60 and 70 kV using Gafchromic® film and an anatomical head phantom. The dose was measured in the occlusal plane, and ± 50 mm cranially and caudally to evaluate the amount of scattered radiation. The same entrance dose to the phantom was used. The absorbed dose was expressed as the ratio of the maximal doses, the mean doses and the integral doses at tube voltages of 70 and 60 kV.ResultsThe patient receives approximately 40 - 50% higher (mean and integral) absorbed dose when a tube voltage of 70 kV is used.ConclusionsThe results of this study clearly indicate that 60 kV should be used for dental intraoral radiographic examinations for approximal caries detection.

A new digital intraoral X-ray image analysis system for objective personal identification

Objectives To develop a new image processing system for personal identification that applies phase-only correlation (POC) to digital intraoral X-ray images.

Forensic radiology in German-speaking area

During the evening of the 8 November in 1895, WilhelmRoentgen, while he was alone in his laboratory at theInstitute of Physics of Julius Maximilians University inWuerzburg, discovered a, so far unknown, kind of radiation.Roentgen called this kind of radiation ‘‘X-rays’’ where the‘‘X’’ stood for the unknown.The famous anatomist Geheimrat von Koelliker thenproposed, on the occasion of a speech for the Physical-Medical Society of Wuerzburg in 1896, to name the X-raysafter its discoverer ‘‘Roentgen-Strahlen’’.The first X-ray images were made on the 22 Novemberwhen Wilhelm Roentgen took a picture of his wife’s (BerthaRoentgen) hand with an exposure time of about 20 min.Many people sensed in those first pictures, which visualizedthe human skeleton, a peculiar preview of death.On 28 December 1895, W. Roentgen submitted his manu-script titled: ‘‘A new kind of radiation’’ to the Physical-Medical Society of Wuerzburg. In that publication, Roent-gen already speaks about a possible application of hisdiscovery in medicine (Figs. 1 and 2).The discovery of Roentgen had an immediate and enor-mous impact, impeding also upon the field of forensicmedicine. Because, just a few weeks later the New YorkSun wrote that Roentgen had already used the new radiationto take pictures of broken bones or projectiles within thebody.InJanuary 1896, theWiener Presse andthe FrankfurterZeitung described a possible clinical application very con-cretely and detailed, e.g. that using the new techniquecomplicated fractures and foreign bodies can be visualizedin a non-invasive manner.Following this, fantasy inspired ideas were born to use thenew X-rays to take pictures of the human soul or of tele-pathic features. Those thoughts reached a maximum, whenan entrepreneur in London, suggested the addition of leadedthreads to women’s underwear to prevent a divesting per-spective using the dreaded ‘‘X-ray-View’’.In 1896 further applications followed and RoentgensX-ray images entered the courtrooms of Northern Americaand England as evidence of bullets within living anddeceased humans [1]. That was the hour of birth of a con-nection between radiology and forensic medicine leading toa clinical–forensic radiology for the documentation of for-ensic findings in the living and to a forensic–pathologicalradiology in the investigation of the deceased.The newly discovered technology in Wuerzburg has alsobeen used to serve as evidence to clarify an assumedmalpractice in the case of an overlooked fracture in Denver(USA) [1].In Munich, Anger had the idea to use an X-ray image ofthe bony structures of the foot for the estimation of age [1].Today, forensic age estimations are still based upon X-rayimages of the joints of the hand, according to Greulich andPyle [2].B. Koenig founded the field of forensic odontology bymaking the first intraoral X-ray images of the teeth [1].The same year Roentgen himself started the industrial,non-destructive, radiological analysis of materials and pro-ducts. As an enthusiastic hunter he additionally made X-raypictures of his shotgun. According his statement, this imageof the shotgun was his best X-ray image. Following this, theDepartment of Defense in Germany and Austria inducted hismethod as an investigation method for weapons [1].Today’s airport luggage scanners are also based onRoentgens invention because it required just 1 year untilthe French customs started to X-ray suspected smugglingpackages [1].Therefore, it is not surprising, that Roentgen received thefirst Nobel Prize in physics of the Royal Swedish Academyof Science on 10 December in 1901. The reclusive Roentgendeclined the invitation to Nobel Prize presentation, and alsothe patent of his discovery. He consigned the Prize of at thattime, 50.000 Krone, to the University of Wuerzburg.Also very early on, during the 1920s, the first proposalsappeared for the application of Roentgen’s method foridentification, namely from A. Schueller. He proposed theuse of a comparison of paranasal sinuses for this [3].

Improvement of the radiation hardness of a directly converting high resolution intra-oral X-ray imaging sensor

The radiation tolerance of a directly converting digital intra-oral X-ray imaging sensor reported in Spartiotis et al. [Nucl. Instr. and Meth. A 501 (2003) 594] has been tested using a typical dental X-ray beam spectrum. Radiation induced degradation in the performance of the sensor which consists of CMOS signal readout circuits bump bonded to a high resistivity silicon pixel detector was observed already before a dose (in air) of 1 krad. Both increase in the leakage current of the pixel detector manufactured by Sintef, Norway and signal leakage to ground from the gate of the pixel input MOSFETs of the readout circuit were observed and measured. The sensitive part of the CMOS circuit was identified as the protection diode of the gate of the input MOSFET. After removing the gate protection diode no signal leakage was observed up to a dose of 5 krad (air) which approximately corresponds to 125.000 typical dental X-ray exposures. The radiation hardness of the silicon pixel detector was improved by using a modified oxidation process supplied by Colibrys, Switzerland. The improved pixel detectors showed no increase in the leakage current at dental doses.

A directly converting high-resolution intra-oral X-ray imaging sensor

A digital intra-oral X-ray imaging sensor with an active area of 3.6×2.9 cm2 and consisting of six charge-integrating CMOS signal readout circuits bump bonded to one high-resistivity silicon pixel detector has been developed and tested. The pixel size is 35 μm. The X-rays entering the sensor window are converted directly to electrical charge in the depleted detector material yielding minimum lateral signal spread and maximum image sharpness. The signal charge is collected on the gates of the input field effect transistors of the CMOS signal readout circuits. The analog signal readout is performed by multiplexing in the current mode independent of the signal charge collection enabling multiple readout cycles with negligible dead time and thus imaging with wide dynamic range. Since no intermediate conversion material of X-rays to visible light is needed, the sensor structure is very compact. The analog image signals are guided from the sensor output through a thin cable to signal processing, AD conversion and memory circuitry, which is realized as an intermediate electronic unit between the sensor and the computer. This unit controls the sensor, collects the data and transfers it via one of the standard computer buses for image processing, display and storage. The overall sensor design and operation are described and experimental MTF, SNR and DQE data are presented together with X-ray images.

Newly designed phantom for visual detection tests of intra-oral x-ray imaging systems

A new phantom set, which consists of ten acrylic plate phantoms with several kinds of holes, was designed for visual detection tests of intra-oral x-ray imagin systems. Detection tests of two kinds of dental x-ray films were performed using three kinds of phantoms, an aluminum test object, Burger's phantom and the new phantom set. Detection-ability curves generated against optical density were compared. Moreover, correlation between the SNR (signal-to-noise ratio) factor and detection ability was analyzed. False positive rates were also compared. To reduce the number of phantom plate that had to be tested in the new phantom set, the number of detected signals for each phantom plate was cumulatively added and statistically compared to the total number of detected signals.

In vitro comparison of subjective image quality of the pana digital intraoral x-ray imaging system and conventional intraoral radiography in caries detection

The aim of this study was to evaluate observer performance with a charge coupled device-based intraoral X-ray imaging system, Pana Digital system, in comparison with conventional film-based radiography (Ektaspeed Plus). A total of 118 teeth mounted in plaster were radiographed using both the Pana Digital system and conventional film. The occlusal and approximal surfaces of the resultant images were evaluated by six observers in relation to the caries depth. For the verification of the caries depth, the teeth surfaces were rated histologically by two examiners according to the same rating scores used for the X-ray imaging assessment. The sensitivity, specificity, positive predictive value, negative predictive value and diagnostic accuracy of the two image modalities were calculated together with Cohen's kappa values. No significant difference was found in any diagnostic parameters at any thresholds between the two modalities. The Kappa values for the Pana Digital system showed relatively lower levels than those of conventional film in occlusal caries assessment, whereas no apparent differences were found between the two modalities in any scores of depth for approximal caries assessment. The results of the present study, in which there was no discrepancy in diagnostic abilities, suggested that the Pana Digital system could play a useful role in clinical caries diagnosis.