Lateral Images Research Articles

Normative vertebral deformity measurements in a clinically relevant population using magnetic resonance imaging.

Osteoporosis is the leading cause of vertebral fractures. Dual-energy X-ray absorptiometry (DXA) and radiographs are traditionally used to detect osteoporosis and vertebral fractures/deformities. Magnetic resonance imaging (MRI) can be utilized to detect the relative severity of vertebral deformities using three-dimensional information not available in traditional DXA and lateral two-dimensional radiography imaging techniques. To generate normative vertebral parameters in women using MRI and DXA scans, determine the correlations between MRI-calculated vertebral deformities and age, DXA T-scores, and DXA Z-scores, and compare MRI vertebral deformity values with radiography values previously published in the literature. This study is a retrospective vertebral morphometric analysis conducted at our institution. The patient sample included MR images from 1638 female patients who underwent both MR and DXA imaging between 2005 and 2014. Biconcavity, wedge, crush, anterior height (Ha)/posterior height (Hp), and middle height (Hm)/posterior height values were calculated from the MR images of the patient's vertebrae. Associations between vertebral deformity values, patient age, and DXA T-scores were analyzed using Spearman correlation. The MRI-derived measurements were compared with radiograph-based calculations from population-based data compiled from multiple studies. Age was positively correlated with lumbar Hm/Hp (P = 0.04) and thoracic wedge (P = 0.03) and biconcavity (P = 0.001) and negatively correlated with thoracic Ha/Hp (P = 0.002) and Hm/Hp (P = 0.001) values. DXA T-scores correlated positively with lumbar Hm/Hp (P < 0.0001) and negatively with lumbar wedge (P = 0.046), biconcavity (P < 0.0001), and Ha/Hp (P = 0.046) values. Qualitative analysis revealed that Ha/Hp differed between MRI and radiography population-based data by no more than 0.3 and Hm/Hp by a maximum of 1.2. Compared with traditional imaging techniques, MRI detects vertebral deformities with high accuracy and reliability. It may be a sensitive, ionizing, radiation-free tool for use in clinical settings.

Comparison between effects of reverse curve of Spee nickel titanium archwire and stainless steel archwires with and without torque on the lower incisors in deep overbite treatment: a randomized control study.

To compare the effect between three different reverse curve of Spee (RCOS) archwires: 0.016 × 0.022-inch Nickel-Titanium and 0.019 × 0.025-inch stainless steel (SS) with and without crown labial torque (CLT) on lower incisors during deep overbite treatment. Eighty subjects with deep overbite were randomly divided into three groups: the first group (mean age: 20.5 years) received SS RCOS with CLT, the second group (mean age: 19.4 years) was treated with 0.019 × 0.025-inch SS RCOS with zero CLT, the third group (mean age: 18.2 years) was treated with rocking-chair nickel-titanium (NiTi) 0.016 × 0.022-inch with RCOS. Two lateral cephalometric images were taken for each patient, one after alignment and the second after deep bite correction. These images were superimposed using the corpus axis to study the lower incisor horizontal and vertical changes. The lower incisor angular change was significantly smaller in Group II (-0.3°) compared to Group I (4.8°) and Group III (6.0°, P ≤ .001). Lower incisor anterior movement was reduced in Group II compared to Group I (P = .014) and Group III (P = .008). Group III showed significantly more downward movement of the lower Incisors (P ≤ .001). The three groups showed comparable amounts of true intrusion (1 mm, P = .536). 0.016 × 0.022-inch NiTi and 0.019 × 0.025-inch SS with crown labial torque RCOS archwires resulted in similar proclination and forward movement of the lower incisors. Removal of anterior crown labial torque from the 0.019 × 0.025-inch SS RCOS archwire prevents lower incisor proclination and forward movement.

Enhancing Sensitivity of Point-of-Care Thyroid Diagnosis via Computational Analysis of Lateral Flow Assay Images Using Novel Textural Features and Hybrid-AI Models.

Lateral flow assays are widely used in point-of-care diagnostics but face challenges in sensitivity and accuracy when detecting low analyte concentrations, such as thyroid-stimulating hormone biomarkers. This study aims to enhance assay performance by leveraging textural features and hybrid artificial intelligence models. A modified Gray-Level Co-occurrence Matrix, termed the Averaged Horizontal Multiple Offsets Gray-Level Co-occurrence Matrix, was utilised to compute the textural features of the biosensor assay images. Significant textural features were selected for further analysis. A deep learning Convolutional Neural Network model was employed to extract features from these textural features. Both traditional machine learning models and hybrid artificial intelligence models, which combine Convolutional Neural Network features with traditional algorithms, were used to categorise these textural features based on the thyroid-stimulating hormone concentration levels. The proposed method achieved accuracy levels exceeding 95%. This pioneering study highlights the utility of textural aspects of assay images for accurate predictive disease modelling, offering promising advancements in diagnostics and management within biomedical research.

Diagnostic value of the airway post-processing technique of multi-slice spiral CT in Accessory cardiac bronchus

Objective: This study aimed to assess the diagnostic value of the airway post-processing technique of multi-slice spiral CT(MSCT) in Accessory cardiac bronchus(ACB)and to improve the understanding of this disease. Methods: The original MSCT axial images and various post-processing reconstructed images of 9 ACB cases were retrospectively analyzed.Airway post-processing techniques, including multi-planar reformation(MPR), minimum intensity projection (MinIP), volume rendering technique(VRT), CT virtual endoscopy(CTVE) and tissue transition projection (TTP), were employed. Three-dimensional recombination images of bronchial trees were obtained and classified according to the Mangiulea classification. Results: Among the 9 cases with ACB, there were 6 cases of accessory-lobed type, 3 cases of long diverticular type, and no cases of short diverticular type. The lateral axial images, MPR, MinIP, VRT, CTVE, and TTP all clearly displayed ACB. Among the 6 cases of accessory-lobed type, there were 5 cases with accessory lobe dysplasia and one case with pneumonia in an accessory lobe;One case was complicated with accessory lobe emphysema. Among the 3 cases with long diverticular type ACB, one had pneumonia. Among the 9 cases, there was 1 case with old tuberculous focus, 1 case with pulmonary fibrous focus, 1 case with pleural thickening and adhesion, and 1 case with tracheal diverticulum. Conclusion: The airway post-processing technique of MSCT can not only clarify the diagnosis, origin and classification of ACB, but also measure the opening diameter, length and distance to the tracheal carina of ACB.It guides clinical treatment and tracheoscopy, etc, and can be used as the preferred examination method for ACB.

The Effect of Selective Dorsal Rhizotomy on Scoliosis in Children With Cerebral Palsy: A Long-term Follow-up Study.

Selective dorsal rhizotomy (SDR) is commonly used to reduce spasticity in children with cerebral palsy (CP). Children with CP have an increased risk of spinal deformities that increase with age and Gross Motor Function Classification System (GMFCS) level. Few studies have considered the risk of spinal deformity post-SDR by GMFCS level. The purpose of this review was to evaluate the incidence and severity of spinal deformities in children with CP, overall and by GMFCS level, a minimum of 5 years after SDR. In this retrospective review, participants included all those who had an SDR before January 1, 2013, at a Canadian pediatric hospital. Participants had to have a spine radiograph preoperatively and a minimum 5 years post-SDR. Age, GMFCS level, level of laminectomy, percentage of rootlets cut, and orthopaedic surgical history were collected. Scoliosis was evaluated using coronal curve and was measured pre-SDR and a minimum 5 years post-SDR or until spine surgery. When available, sagittal plane radiographs were reviewed. A total of 107 participants (61 male, 46 female), at GMFCS levels I to V (2,25,22,45, and 13, respectively) were included. The mean age at SDR was 4.8 years (SD 1.5), with a mean follow-up of 12.7 years (SD 4.9). Post-SDR, 57 (53%) participants had scoliosis, with the major curve averaging 46 degrees (SD 34.9). Scoliosis measuring >10 degrees was found in 8 (32%) participants at level II, 4 (18%) at level III, 33 (73%) at level IV and 12 (92%) at level V. No participants at GMFCS levels I to III, and 35% and 67% of those at levels IV and V, respectively, had a curve magnitude >40 degrees. Of those who had lateral imaging, 38% had excessive kyphosis, 21% hyperlordosis, and 16% spondylolisthesis. The rate of scoliosis is consistent with the natural history of children with CP without surgical intervention as reported in the literature when compared by GMFCS level and curve severity. Level IV.

Evaluation of the true lateral fluoroscopic projection for the relation of the S1 recess/foramen to safe corridors in transiliac-transsacral screw placement in human cadaveric pelves

IntroductionPercutaneous screw fixation is a widely used treatment for posterior pelvic ring injuries. Transiliac-transsacral screw fixation has demonstrated superior biomechanical properties over bilateral sacroiliac screws, particularly in the minimally displaced bilateral sacral fractures. Screw placement under fluoroscopic control is still common, while CT navigation is gaining popularity. However, the accurate placement of screws within a safe zone is essential to avoid neurovascular complications.MethodsAn anatomical study using human cadaveric pelves was conducted to assess radiological landmarks and determine a safe zone in relation to the S1 recess/foramen for transiliac-transsacral screw placement.ResultsFourteen pelves were evaluated. Ten pelves were classified as having a satisfactory corridor for screw placement, while four were deemed to have an impossible or high-risk corridor. A safe zone was defined based on the diagonal bisector of the S1 vertebral body, ICD and anterior cortex.DiscussionThe study findings suggest that lateral fluoroscopic projection can determine a safe entry point for screw placement. Understanding the anatomy and landmarks on lateral fluoroscopic images is crucial for successful screw placement and avoiding complications.ConclusionThe S1 body diagonal is consistently located anterior to the S1 recess in lateral fluoroscopic projections, providing a potential safe corridor for transiliac-transsacral screw placement at the S1 level in nondysmorphic pelves. Further research is needed to confirm these findings with CT imaging and evaluate the technical feasibility of screw placement.

Development of a deep learning model for automatic detection of narrowed intervertebral disc space sites in caudal thoracic and lumbar lateral X-ray images of dogs.

Intervertebral disc disease is the most common spinal cord-related disease in dogs, caused by disc material protrusion or extrusion that compresses the spinal cord, leading to clinical symptoms. Diagnosis involves identifying radiographic signs such as intervertebral disc space narrowing, increased opacity of the intervertebral foramen, spondylosis deformans, and magnetic resonance imaging findings like spinal cord compression and lesions, alongside clinical symptoms and neurological examination findings. Intervertebral disc space narrowing on radiographs is the most common finding in intervertebral disc extrusion. This study aimed to develop a deep learning model to automatically recognize narrowed intervertebral disc space on caudal thoracic and lumbar X-ray images of dogs. In total, 241 caudal thoracic and lumbar lateral X-ray images from 142 dogs were used to develop and evaluate the model, which quantified intervertebral disc space distance and detected narrowing using a large-kernel one-dimensional convolutional neural network. When comparing veterinary clinicians and the deep learning model, the kappa value was 0.780, with 81.5% sensitivity and 95.6% specificity, showing substantial agreement. In conclusion, the deep learning model developed in this study, automatically and accurately quantified intervertebral disc space distance and detected narrowed sites in dogs, aiding in the initial screening of intervertebral disc disease and lesion localization.

A deep learning based automatic two-dimensional digital templating model for total knee arthroplasty

BackgroundPreoperative templating is an important step for total knee arthroplasty (TKA), facilitating hospital personnel in the anticipation and preparation of necessary surgical resources. Despite its importance, this process currently lacks automation. This study aimed to develop an artificial intelligence (AI) model to automate implant size prediction.MethodsA total of 13,281 (2938 anteroposterior, 10,343 lateral) knee radiographs obtained from the authors’ institute were utilized for model training, with 2302 (1034 anteroposterior, 1268 lateral) images set apart for validation and testing. The templating AI model integrates a pipeline composed of multiple steps for automated implant size estimation. To predict implant size, anterioposterior (AP) and lateral radiograph predictions were merged, selecting the smaller of the predicted sizes to prevent implant overhang. The model’s size predictions were validated with 81 real TKA data set apart from the training data, and its accuracy was compared to that of manual templating by an orthopedic specialist. Predictions matching the actual implanted sizes were labeled “exact” and those within one size, “accurate.” The influence of patient characteristics on the model’s prediction accuracy was also analyzed. The measurement time elapsed for implant sizing was recorded for both the AI model and the orthopedic specialist. Implant position predicted by the model was validated by comparing insert locations with postoperative images.ResultsCompared with data from 81 actual TKA procedures, the model provided exact predictions for 39.5% of femoral and 43.2% of tibial components. Allowing a one-size margin of error, 88.9% of predictions were deemed “accurate” for both components. Interobserver reliability (Cohen’s kappa) were 0.60 and 0.70 for femoral and tibial implants, respectively, both classified as “substantial.” The orthopedic specialist produced results accurate within one-size margin of error in 95.1% and 100% of cases for femoral and tibial components, respectively. Interobserver reliability between the orthopedic specialist and ground truth was 0.76 and 0.8 for femoral and tibial components, respectively. The measurement time per case was 48.7 s for the AI model, compared with 97.5 s for the orthopedic specialist. Compared with postoperative radiographs, predicted implant position had an error of less than 4 mm on average.ConclusionsAn AI-based templating tool for TKA was successfully developed, demonstrating satisfactory accuracy and efficiency. Its application could significantly reduce the clinical workload in TKA preparation.

(Invited) High-Performance Ge/Si Electro-Absorption Optical Modulator and Its Highly-Efficient Photodetector Operation

We studied a high-speed Ge/Si EAM evanescently coupled with a Si waveguide of a p–n junction for a high-bandwidth optical interconnect over a wide range of temperatures from 25 °C to 85 °C.Figure 1 (a) -(c) show schematic diagrams of a Ge/Si EAM on a Si rib waveguide with a lateral pn junction and cross-sectional TEM image of a Ge/Si EAM. The Ge/Si EAM consists of an evanescently coupled structure with a Si rib waveguide. The stack Ge/Si layer width was changed from 0.3 mm to 1.0 mm, and the Ge height was about 300 nm.Figure 2 (a) and (b) show output waveforms of 112 Gbps PAM-4 with 2^7-1 PRBS and 100 Gbps NRZ with 2^7-1 PRBS at 1550 nm wavelength, measured at room temperature.We demonstrated 56 Gbps high-speed operation at high temperatures up to 85 °C. From the photoluminescence spectra, we confirmed that the bandgap energy dependence on temperature is relatively small, which is consistent with that of the operation wavelengths with increasing temperature for a Ge/Si EAM. We also demonstrate a high-speed and high-efficiency Ge photodetector. The photoresponsivity was enhanced by the F-K effect and avalanche-multiplication effect in the C+L band wavelengths. These results demonstrate that the Ge/Si stacked structure is promising for both high-performance optical modulators and photodetectors integrated on Si platforms. Reference: J. Fujikata et al., Opt. Express 31(6), 10732 (2023).J. Fujikata et al., Opt. Express 28(22), 33123 (2020). Figure 1

Accuracy Evaluation of a Three-Dimensional Face Reconstruction Model Based on the Hifi3D Face Model and Clinical Two-Dimensional Images.

Three-dimensional (3D) facial models have been increasingly applied in orthodontics, orthognathic surgery, and various medical fields. This study proposed an approach to reconstructing 3D facial models from standard orthodontic frontal and lateral images, providing an efficient way to expand 3D databases. A total of 23 participants (average age 20.70 ± 5.36 years) were enrolled. Based on the Hifi3D face model, 3D reconstructions were generated and compared with corresponding face scans to evaluate their accuracy. Root mean square error (RMSE) values were calculated for the entire face, nine specific facial regions, and eight anatomical landmarks. Clinical feasibility was further assessed by comparing six angular and thirteen linear measurements between the reconstructed and scanned models. The RMSE of the reconstruction model was 2.00 ± 0.38 mm (95% CI: 1.84-2.17 mm). High accuracy was achieved for the forehead, nose, upper lip, paranasal region, and right cheek (mean RMSE < 2 mm). The forehead area showed the smallest deviation, at 1.52 ± 0.88 mm (95% CI: 1.14-1.90 mm). In contrast, the lower lip, chin, and left cheek exhibited average RMSEs exceeding 2 mm. The mean deviation across landmarks was below 2 mm, with the Prn displaying the smallest error at 1.18 ± 1.10 mm (95% CI: 0.71-1.65 mm). The largest discrepancies were observed along the Z-axis (Z > Y > X). Significant differences (p < 0.05) emerged between groups in the nasolabial, nasal, and nasofrontal angles, while the other 13 linear and 3 angular measurements showed no statistical differences (p > 0.05). This study explored the feasibility of reconstructing accurate 3D models from 2D photos. Compared to facial scan models, the Hifi3D face model demonstrated a 2 mm deviation, with potential for enriching 3D databases for subjective evaluations, patient education, and communication. However, caution is advised when applying this model to clinical measurements, especially angle assessments.

Comparative efficacy of anteroposterior and lateral X-ray based deep learning in the detection of osteoporotic vertebral compression fracture

Magnetic resonance imaging remains the gold standard for diagnosing osteoporotic vertebral compression fractures (OVCF), but the use of X-ray imaging, particularly anteroposterior (AP) and lateral views, is prevalent due to its accessibility and cost-effectiveness. We aim to assess whether the performance of AP images-based deep learning is comparable compared to those using lateral images. This retrospective study analyzed X-ray images from two tertiary teaching hospitals, involving 1,507 patients for the training and internal test, and 104 patients for the external test. The EfficientNet-B5-based algorithms were employed to classify OVCF and non-OVCF group. The model was trained with a 1:1 balanced dataset and validated through 5-fold cross validation. Performance outcomes were compared with the area under receiver operating characteristic (AUROC) curve. Out of a total of 1,507 patients, 799 were included in the training dataset and 708 were included in the internal test dataset. The training and internal test datasets were matched 1:1 as OVCF and non-OVCF patients. The DL model showed comparable classifying performance with internal test data (N = 708, AUROC for AP, 0.915; AUROC for lateral, 0.953) and external test data (N = 104, AUROC for AP, 0.982; AUROC for lateral, 0979), respectively. The other performances including F1 score and accuracy were also comparable. Especially, The AUROC of AP and lateral x-ray image-based DL was not significantly different (p for DeLong test = 0.604). The EfficientNet-B5 algorithms using AP X-ray images shows comparable efficacy for classifying OVCF and non-OVCF compared to lateral images.

Optimization and application of renal depth measurement method in the cadmium-zinc-telluride‑based SPECT/CT renal dynamic imaging

PurposeThis study aims to evaluate the accuracy of four kidney depth measurement methods—nuclear medicine tomography, nuclear medicine lateral scanning, ultrasound, and Tonnesen’s formula-based estimation—using CT measurements as the reference standard. Additionally, it investigates the feasibility of utilizing nuclear medicine tomography and lateral scanning for measuring kidney depth in 99mTc-DTPA renal dynamic imaging.MethodsHollow kidney phantoms mimicking the shape and volume of adult kidneys were 3D printed and filled with 99mTcO4− solution. These phantoms were then subjected to lateral scanning and nuclear medicine tomography using CZT (cadmium-zinc-telluride) SPECT/CT to determine the optimal post-processing method. Forty patients who underwent renal dynamic imaging were recruited for the study. Renal depths were derived from ultrasound, lateral imaging, nuclear medicine tomography, formula-based estimation, and CT measurements. The renal depths obtained through these four methods were for correlation with CT-measured renal depths. Additionally, the absolute differences between renal depths obtained by each method and the CT standard were analyzed and compared across groups.ResultsUsing kidney phantoms, nuclear medicine tomography images were processed with a Butterworth filter (cutoff frequency = 0.6), and renal outlines in lateral images was manually delineated. In the clinical validation phase, correlation coefficients indicated strong associations between renal depths measured by nuclear medicine tomography (left kidney: R = 0.885, P < 0.05; right kidney: R = 0.927, P < 0.05) and lateral scanning (left kidney: R = 0.933, P < 0.05; right kidney: R = 0.956, P < 0.05) compared to CT measurements. The difference in kidney depth between nuclear medicine tomography and CT measurements were the smallest and statistically significant (left kidney: 0.69 ± 0.51; right kidney: 0.58 ± 0.41, P < 0.05).ConclusionUsing ordered subset expectation maximization (OSEM) in conjunction with a Butterworth filter (fc = 0.6) as the post-processing method, nuclear medicine tomography enables more accurate renal depth measurements without increasing the radiation dose to patients.

Systematic evaluation of abdominal aortic calcification in patients with a recent clinical fracture visiting the Fracture Liaison Service.

Abdominal aortic calcification (AAC) is associated with an increased risk of cardiovascular disease (CVD), osteoporosis and fractures. We aimed to analyze the prevalence and severity of AAC and to assess whether index fracture location, bone mineral density (BMD) and prevalent VFs are associated with AAC in patients with a recent fracture. Cross-sectional cohort study of patients with a recent clinical fracture (aged 50-90years) attending the FLS. Patients received a BMD measurement and lateral spine imaging using Dual-energy X-ray absorptiometry. AAC prevalence was assessed using the AAC-24 score and categorized as none, moderate (AAC-24 1-4) and severe (AAC-24 ≥ 5). Multivariate logistic regression analyses were performed to study the association between risk factors and AAC presence/ severity. AAC was present in 478 (27.6%) of 1731 patients of whom 207 (43.3%) had moderate and 271 (56.7%) severe AAC. The presence of AAC was associated with age, BMI, smoking, history of CVD and prevalent grade 2 or 3 VFs, but index fracture location and BMD were not associated with AAC or severe AAC. In patients with AAC (n = 318) without a history of CVD, there was no association between index fracture location and BMD. In that subgroup, severe AAC was not associated with prevalent VFs. In FLS patients, the prevalence of AAC and severe AAC was 27.6% and 15.7%. Index fracture location and BMD were not associated with AAC or severe AAC. Prevalent VFs were associated with AAC, but not with severe AAC in the subgroup of patients without CVD.

Efficacy of additional lateral pinhole and SPECT/CT imaging in dual-phase Tc-99m MIBI parathyroid scintigraphy for localising parathyroid pathologies in patients with primary hyperparathyroidism: a single-institution experience.

Parathyroid imaging with dual-phase technetium-99m methoxyisobutrylizonitrile (Tc-99m MIBI) scintigraphy serves as an important prerequisite for the identification of hyperfunctioning parathyroid gland(s) in patients with primary hyperparathyroidism (PHPT) for a successful targeted parathyroidectomy. This study aimed to evaluate the clinical value of additional lateral imaging and single-photon emission computed tomography/computed tomography (SPECT/CT) versus conventional planar imaging for locating parathyroid pathologies in patients with PHPT. A retrospective review was performed on 105 patients who underwent dual-phase Tc-99m MIBI scintigraphy and were surgically treated by parathyroidectomy. Dual-phase Tc-99m-MIBI planar scintigraphy with additional lateral pinhole views and SPECT/CT imaging was performed on a routine basis, as per departmental protocol. Comparison study between imaging modalities was done by patient-based analysis and scintigraphy results were compared with the clinical findings, biochemical markers, and histopathological findings. Sensitivity and specificity for anterior planar dual-phase Tc-99m MIBI scintigraphy were 78.8 and 80%, respectively. In comparison, lateral pinhole scan and SPECT/CT alone were found to have sensitivities of 85.9 and 90.9%, respectively, with the same specificity. Sensitivity decreased in patients with normocalcaemia and multiglandular disease. The mean adenoma weight and size for true-positive studies were significantly higher than those for false-negative or false-positive studies. SPECT/CT provided the highest diagnostic accuracy for preoperative identification of parathyroid lesions in PHPT patients. Lateral pinhole imaging offers comparable sensitivity and aids in adenoma localisation when SPECT/CT is unavailable.

Synthesis and Characterization of Boron Nitride Thin Films Deposited by High-Power Impulse Reactive Magnetron Sputtering.

In the present research, hexagonal boron nitride (h-BN) films were deposited by reactive high-power impulse magnetron sputtering (HiPIMS) of the pure boron target. Nitrogen was used as both a sputtering gas and a reactive gas. It was shown that, using only nitrogen gas, hexagonal-boron-phase thin films were synthesized successfully. The deposition temperature, time, and nitrogen gas flow effects were studied. It was found that an increase in deposition temperature resulted in hydrogen desorption, less intensive hydrogen-bond-related luminescence features in the Raman spectra of the films, and increased h-BN crystallite size. Increases in deposition time affect crystallites, which form larger conglomerates, with size decreases. The conglomerates' size and surface roughness increase with increases in both time and temperature. An increase in the nitrogen flow was beneficial for a significant reduction in the carbon amount in the h-BN films and the appearance of the h-BN-related features in the lateral force microscopy images.

Compare three deep learning-based artificial intelligence models for classification of calcified lumbar disc herniation: a multicenter diagnostic study.

To develop and validate an artificial intelligence diagnostic model for identifying calcified lumbar disc herniation based on lateral lumbar magnetic resonance imaging(MRI). During the period from January 2019 to March 2024, patients meeting the inclusion criteria were collected. All patients had undergone both lumbar spine MRI and computed tomography(CT) examinations, with regions of interest (ROI) clearly marked on the lumbar sagittal MRI images. The participants were then divided into separate sets for training, testing, and external validation. Ultimately, we developed a deep learning model using the ResNet-34 algorithm model and evaluated its diagnostic efficacy. A total of 1,224 eligible patients were included in this study, consisting of 610 males and 614 females, with an average age of 53.34 ± 10.61 years. Notably, the test datasets displayed an impressive classification accuracy rate of 91.67%, whereas the external validation datasets achieved a classification accuracy rate of 88.76%. Among the test datasets, the ResNet34 model outperformed other models, yielding the highest area under the curve (AUC) of 0.96 (95% CI: 0.93, 0.99). Additionally, the ResNet34 model also exhibited superior performance in the external validation datasets, exhibiting an AUC of 0.88 (95% CI: 0.80, 0.93). In this study, we established a deep learning model with excellent performance in identifying calcified intervertebral discs, thereby offering a valuable and efficient diagnostic tool for clinical surgeons.

Predicting OSA Using Radiographs of the Airway Anatomy.

This study aims to analyze the tongue body shape and upper airway anatomical parameters in patients with Obstructive Sleep Apnea (OSA) and to explore the anatomical causes of OSA. A total of 345 subjects participated in this study. Lateral pharyngeal images of the upper respiratory tract were captured in both normal and mandibular advancement states using X-ray plain film. Measurements were taken for the following parameters: Tongue Length, Tongue Thickness, Distance from the Mandibular Plane to the Hyoid, Soft Palate Length, Posterior Oropharyngeal Depth, Palatal Airway Space, Tongue Depth Space, and Mental Posterior Space. The correlation between the Apnea-Hypopnea Index (AHI) and these upper airway anatomical factors was analyzed using both univariate and multivariate analyses to develop a predictive model for OSA. The anatomical structure of the upper airway in patients with OSA is narrower compared to non-OSA individuals, and these patients exhibit a longer and thicker tongue. During mandibular advancement, the pharyngeal airway widens; however, the tongue length decreases while its thickness increases. Univariate correlation analysis revealed that the severity of OSA was significantly associated with tongue length, the ratio of tongue length to tongue thickness, the distance from the mandibular plane to the hyoid, soft palate length, and body mass index (BMI) in both the normal position and during mandibular advancement (p < 0.001). Multivariate linear analysis indicated that the severity of OSA is linked with the mandibular plane to hyoid distance in the normal position (MPH(N)) and BMI. A nomogram was utilized to develop a predictive model for OSA, achieving an area under the receiver operating characteristic curve of 0.838. The pathogenesis of OSA is related to pharyngeal anatomy and tongue length in the state of mandibular advancement, which can be predicted by the measurement indexes of normal and anterior mandibular displacement lateral pharyngeal radiograph. This may potentially aid in early screening and diagnosis of OSA.

Role of Artificial Intelligence in Determining Ideal Nasal Tip Projection in Diverse Populations.

Nasal tip projection ideals have been based on the Caucasian nose; however, many racial groups have underprojected tips. Several methodologies have been described to measure ideal nasal tip projection. This study aimed to compare various methods to racially diverse groups to elucidate those best suited for race-specific facial morphology. An openly accessible artificial intelligence image generator was used to generate images of Caucasian, Middle Eastern, and African American noses. Tip projection was manually measured using 5 different ideal tip projection methods. One-way analysis of variance with post hoc Tukey honest significant difference test evaluated differences between racial cohorts. One-sample t tests were used for comparisons to previously described aesthetic ideals, and a P value of less than 0.05 was considered statistically significant. Twenty lateral images were analyzed per racial cohort (10 men and 10 women). All racial groups-except for male Middle Eastern noses (P = 0.01)-were measured as ideal using the Byrd nasal length proportion method. Caucasian and Middle Eastern noses measured ideal using the nasolabial tip projection method. African American female noses measured ideal using the nasofacial tip projection method. Only Caucasian noses were ideal using the lip tangent method. This study suggests that the ideal nasal length to tip length ratio of two-thirds (67%) has generalizability across these racial groups; the lip tangent is an excellent method for evaluating tip projection in Caucasians; and assessment of ideal tip projection in Middle Eastern and African American groups should use regional midface and lower face anatomy to account for differences in facial morphology.

Enhancing Diagnostic Accuracy of Fresh Vertebral Compression Fractures with Deep Learning Models.

Retrospective study. The study aimed to develop and authenticated a deep learning model based on X-ray images to accurately diagnose fresh thoracolumbar vertebral compression fractures. In clinical practice, diagnosing fresh vertebral compression fractures often requires MRI. However, due to the scarcity of MRI resources and the high time and economic costs involved, some patients may not receive timely diagnosis and treatment. Using a deep learning model combined with X-rays for diagnostic assistance could potentially serve as an alternative to MRI. In this study, the main collection included X-ray images suspected of thoracolumbar vertebral compression fractures from the municipal shared database between December 2012 and February 2024. Deep learning models were constructed using frameworks of EfficientNet, MobileNet, and MnasNet respectively. We conducted a preliminary evaluation of the deep learning model using the validation set. The diagnostic performance of the models was evaluated using metrics such as AUC value, accuracy, sensitivity, specificity, F1 score, precision, and ROC curve. Finally, the deep learning models were compared with evaluations from two spine surgeons of different experience levels on the control set. This study included a total of 3025 lateral X-ray images from 2224 patients. The dataset was divided into a training set of 2388 cases, a validation set of 482 cases, and a control set of 155 cases. In the validation set, the three groups of DL models had accuracies of 83.0%, 82.4%, and 82.2%, respectively. The AUC values were 0.861, 0.852, and 0.865, respectively. In the control set, the accuracies of the three groups of DL models were 78.1%, 78.1%, and 80.7%, respectively, all higher than spinal surgeons and significantly higher than junior spine surgeon. This study developed deep learning models for detecting fresh vertebral compression fractures, demonstrating high accuracy.

Measurement Method of Femoral Anteversion During Surgery for Trochanteric Fractures.

The acceptable range of bilateral differences in the femoral rotation is ≤15°, as a difference in femoral rotation over 15° may cause functional disturbance. However,femoral trochanteric fractures do not have clear indicators of rotation during surgery. This study aimed to verify the accuracy of this novel method for measuring femoral anteversion during trochanteric fracture surgery. From August 2022 to August 2024, this study prospectively included patients with femoral trochanteric fractures treated using a cephalo-medullary nail with a lag-screw neck-shaft angle of 125°. During surgery, direct intraoperative anteversion (DIAV) was measured using the hip post axis in the lateral hip image based on a tabletop plane parallel to the floor. DIAV was then converted to corrected intraoperative anteversion (CIAV) using a quick chart derived from a graph of the axial projection. The accuracy was analyzed by comparing CIAV with computed tomography anteversion (CTAV) measurements obtained after surgery. One hundred patients (25 male, 75 female) with a mean age of 86.7 years (range: 67-104 years) were included in this study. According to the Orthopaedic Trauma Association classification, there were 66 patients classified as A1, 23 as A2, and 11 as A3. The mean DIAV and CIAV were 8.2°±8.2° and 9.6°±9.7°, respectively, while the mean CTAV was 10.3°±10.4°. The median difference between CIAV and CTAV was 3° (range: 0°-9°), with 84 patients exhibiting differences of ≤5°. No significant differences were found between CIAV and CTAV (p = 0.054), whereas DIAV was significantly lower than CTAV (p < 0.001). CIAV and CTAV were strongly correlated (r = 0.936, p < 0.001). The Bland-Altman plot between CIAV and CTAV revealed that 98 patients were within the limits of agreement, and the plot distribution showed no trend. The measurement method for assessing femoral anteversion using the hip post axis on the lateral hip image based on a tabletop plane parallel to the floor during surgery demonstrated sufficient accuracy for indicating anteversion in femoral trochanteric fractures.