Search

Abstract Background The Kellgren-Lawrence (KL) grading system is the most widely used method to classify the severity of osteoarthritis (OA) of the knee. However, due to ambiguity of terminology, the KL system showed inferior inter- and intra-observer reliability. For a more reliable evaluation, we recently developed novel deep learning (DL) software known as MediAI-OA to extract each radiographic feature of knee OA and to grade OA severity based on the KL system. Methods This research used data from the Osteoarthritis Initiative for training and validation of MediAI-OA. 44,193 radiographs and 810 radiographs were set as the training data and used as validation data, respectively. This AI model was developed to automatically quantify the degree of joint space narrowing (JSN) of medial and lateral tibiofemoral joint, to automatically detect osteophytes in four regions (medial distal femur, lateral distal femur, medial proximal tibia and lateral proximal tibia) of the knee joint, to classify the KL grade, and present the results of these three OA features together. The model was tested by using 400 test datasets, and the results were compared to the ground truth. The accuracy of the JSN quantification and osteophyte detection was evaluated. The KL grade classification performance was evaluated by precision, recall, F1 score, accuracy, and Cohen's kappa coefficient. In addition, we defined KL grade 2 or higher as clinically significant OA, and accuracy of OA diagnosis were obtained. Results The mean squared error of JSN rate quantification was 0.067 and average osteophyte detection accuracy of the MediAI-OA was 0.84. The accuracy of KL grading was 0.83, and the kappa coefficient between the AI model and ground truth was 0.768, which demonstrated substantial consistency. The OA diagnosis accuracy of this software was 0.92. Conclusions The novel DL software known as MediAI-OA demonstrated satisfactory performance comparable to that of experienced orthopedic surgeons and radiologists for analyzing features of knee OA, KL grading and OA diagnosis. Therefore, reliable KL grading can be performed and the burden of the radiologist can be reduced by using MediAI-OA.

This study aimed to investigate whether the left and right hands yield the same bone age using the automated bone age assessment (BAA) system and proposed the right-hand BAA as a tool for filtering out measurement errors. The Bland–Altman, Passing–Bablok, and Spearman correlation coefficients were analyzed to compare the automated BAA results for each hand. The absolute difference between each hand obtained by the model (ADBH model) was calculated. The mean absolute difference (MAD) was estimated between the automatic BAA results for each hand and the reference standard. The mean of the ADBH model was 0.23 ± 0.19 years; 92.2% of the participants showed an ADBH model result of 0.5 years, with a MAD between the model and the reference standard of 0.409 years for the left hand and 0.424 years for the right hand; both MADs were higher than those of previous studies using the same model. Given the excellent overall correlation of the BAA between both hands using the model, the high ADBH model value may indicate BAA measurement errors and serve as a cue for manual supervision.

Abstract The present study aimed to evaluate the performance of automated skeletal maturation assessment system for Fishman’s skeletal maturity indicators (SMI) for the use in dental fields. Skeletal maturity is particularly important in orthodontics for the determination of treatment timing and method. SMI is widely used for this purpose, as it is less time-consuming and practical in clinical use compared to other methods. Thus, the existing automated skeletal age assessment system based on Greulich and Pyle and Tanner-Whitehouse3 methods was further developed to include SMI using artificial intelligence. This hybrid SMI-modified system consists of three major steps: (1) automated detection of region of interest; (2) automated evaluation of skeletal maturity of each region; and (3) SMI stage mapping. The primary validation was carried out using a dataset of 2593 hand-wrist radiographs, and the SMI mapping algorithm was adjusted accordingly. The performance of the final system was evaluated on a test dataset of 711 hand-wrist radiographs from a different institution. The system achieved a prediction accuracy of 0.772 and mean absolute error and root mean square error of 0.27 and 0.604, respectively, indicating a clinically reliable performance. Thus, it can be used to improve clinical efficiency and reproducibility of SMI prediction.

Background: Radiographs are widely used to evaluate radiographic progression with modified stoke ankylosing spondylitis spinal score (mSASSS). Objective: This pilot study aimed to develop a deep learning model for grading the corners of the cervical and lumbar vertebral bodies for computer-aided detection of mSASSS in patients with ankylosing spondylitis (AS). Methods: Digital radiographic examination of the spine was performed using Discovery XR656 (GE Healthcare) and Digital Diagnost (Philips). The disk points were detected between the bodies using a key-point detection deep learning model from the image obtained in DICOM (digital imaging and communications in medicine) format from the cervical and lumbar spinal radiographs. After cropping the vertebral regions around the disk point, the lower and upper corners of the vertebral bodies were classified as grade 3 (total bony bridges) or grades 0, 1, or 2 (non-bridges). We trained a convolutional neural network model to predict the grades in the lower and upper corners of the vertebral bodies. The performance of the model was evaluated in a validation set, which was separate from the training set. Results: Among 1280 patients with AS for whom mSASSS data were available, 5,083 cervical and 5245 lumbar lateral radiographs were reviewed. The total number of corners where mSASSS was measured in the cervical and lumbar vertebrae, including the upper and lower corners, was 119,414. Among them, the number of corners in the training and validation sets was 110,088 and 9326, respectively. The mean accuracy, sensitivity, and specificity for mSASSS scoring in one corner of the vertebral body were 0.91604, 0.80288, and 0.94244, respectively. Conclusion: A high-performance deep learning model for grading the corners of the vertebral bodies was developed for the first time. This model must be improved and further validated to develop a computer-aided tool for assessing mSASSS in the future.

Objective To evaluate correlations between values of articulation tests and language tests for children with articulation disorder in Korea. Methods Data of outpatients with chief complaint of an articulation problem were retrospectively collected. Patients who underwent Urimal Test of Articulation and Phonation (U-TAP) with Assessment of Phonology and Articulation for Children (APAC), Preschool Receptive-Expressive Language Scale (PRES), or Receptive and Expressive Vocabulary Test (REVT) simultaneously were identified. Patients whose word-level percentages of correct consonants in U-TAP (UTAP_wC) were more than 2 standard deviations below the mean as diagnostic criteria for articulation disorder were selected. Those whose receptive language age (P_RLA), expressive language age (P_ELA), or combined language age (P_CLA) in PRES was delayed more than 24 months compared to their chronological age in months as diagnostic criteria for language disorder were excluded. Results Thirty-three children aged 3–6 years were enrolled retrospectively. PRES and U-TAP showed significant correlations for most of value relationships. PRES and APAC showed significant correlations for all value relationships except for receptive language age. All values of REVT were significantly correlated with all values from U-TAP, but not with any value from APAC. Articulation tests U-TAP and APAC showed significant correlations between percentages of correct consonants. Language tests PRES and REVT showed significant correlations for all value relationships. Conclusion This study suggests that articulation abilities and language abilities might be correlated in children with articulation disorder.

Prior studies examining predictors of favorable clinical outcomes after upper limb robot-assisted therapy (RT) have many shortcomings. Therefore, the aim of this study was to identify meaningful predictors and a prediction model for clinically significant motor improvement in upper limb impairment after RT for each stroke phase. This retrospective, single-center study enrolled patients with stroke who received RT using InMotion2 along with conventional therapy (CT) from January 2015 to September 2019. Demographic characteristics, clinical measures, and robotic kinematic measures were evaluated. The primary outcome measure was the Fugl-Meyer Assessment-Upper Extremity (FMA-UE) and we classified patients with improvement more than the minimal clinically important difference as responders for each stroke phase. Univariable and multivariable logistic regression analyses were performed to assess the relationship between potential predictors and RT responders and determine meaningful predictors. Subsequently, meaningful predictors were included in the final prediction model. One hundred forty-four patients were enrolled. The Hand Movement Scale and time since onset were significant predictors of clinically significant improvement in upper limb impairment (P = 0.045 and 0.043, respectively), as represented by the FMA-UE score after RT along with CT, in patients with subacute stroke. These variables were also meaningful predictors with borderline statistical significance in patients with chronic stroke (P = 0.076 and 0.066, respectively). Better hand movement and a shorter time since onset can be used as realistic predictors of clinically significant motor improvement in upper limb impairment after RT with InMotion2 alongside CT in patients with subacute and chronic stroke. This information may help healthcare professionals discern optimal patients for RT and accurately inform patients and caregivers about outcomes of RT.

Objectives Electrocardiograhy (ECG) is the first step in hypertrophic cardiomyopathy (HCMP) diagnosis. For various reasons, the T wave inversion (TWI) and ECG change with time and HCMP is not easy to diagnosis. The aim of this retrospective study was to investigate the association between TWI on ECG and apical HCMP. Methods A total of 4,730 ECGs presenting TWI from January 2011 to March 2013 in Pusan National University Hospital were enrolled. 133 patients who were examined by both echocardiography and coronary angiogram were analyzed. Patients were divided into two groups: Group A (TWI ≥ 10 mm) and Group B (5 mm ≤ TWI < 10 mm). HCMP is defined by a wall thickness ≥ 15mm in one or more LV myocardial segments. Apical HCMP is defined to be hypertrophy that is confined to LV apex. The patients who had ECGs with at least one month interval were divided 3 groups: Normal T wave, Abnormal T wave, and Persistent TWI. The prevalence of Apical HCMP and coronary artery disease (CAD) was reviewed among the three groups. Results In this study there were a total 133 patients, with patients divided into Group A which had 15 patients and Group B which had 118 patients. Among the 23 patients with apical HCMP, three patients were Group A and twenty patients were Group B (P = 0.769). Regarding constancy of TWI, persistent TWI group was higher in apical HCMP than in other groups (P = 0.038). CAD had no difference between groups (P = 0.889). Conclusions T wave negativity was not associated with incidence of apical HCMP. However, apical HCMP was diagnosed more frequently in patients with persistent TWI. Further follow up echocardiographic study is needed to evaluate the progression of apical HCMP in patients with TWI.

Objective This study was conducted to enable on-line prediction of primal and commercial cut weights in Korean slaughter pigs by AutoFom III, which non-invasively scans pig carcasses early after slaughter using ultrasonic sensors. Methods A total of 162 Landrace, Yorkshire, and Duroc (LYD) pigs and 154 LYD pigs representing the yearly Korean slaughter distribution were included in the calibration and validation dataset, respectively. Partial least squares (PLS) models were developed for prediction of the weight of deboned shoulder blade, shoulder picnic, belly, loin, and ham. In addition, AutoFom III’s ability to predict the weight of the commercial cuts of spare rib, jowl, false lean, back rib, diaphragm, and tenderloin was investigated. Each cut was manually prepared by local butchers and then recorded. Results The cross-validated prediction accuracy (R2cv) of the calibration models for deboned shoulder blade, shoulder picnic, loin, belly, and ham ranged from 0.77 to 0.86. The R2cv for tenderloin, spare rib, diaphragm, false lean, jowl, and back rib ranged from 0.34 to 0.62. Because the R2cv of the latter commercial cuts were less than 0.65, AutoFom III was less accurate for the prediction of those cuts. The root mean squares error of cross validation calibration (RMSECV) model was comparable to the root mean squares error of prediction (RMSEP), although the RMSECV was numerically higher than RMSEP for the deboned shoulder blade and belly. Conclusion AutoFom III predicts the weight of deboned shoulder blade, shoulder picnic, loin, belly, and ham with high accuracy, and is a suitable process analytical tool for sorting pork primals in Korea. However, AutoFom III’s prediction of smaller commercial Korean cuts is less accurate, which may be attributed to the lack of anatomical reference points and the lack of a good correlation between the scanned area of the carcass and those traits.

Temperature is a fundamental physical quantity whose accurate measurement is of critical importance in virtually every area of science, engineering, and biomedicine. Temperature can be measured in many ways. In this pedagogically focused review, we briefly discuss various standard contact thermometry measurement techniques. We introduce and touch upon the necessity of non-contact thermometry, particularly for systems in extreme environments and/or in rapid motion, and how luminescence thermometry can be a solution to this need. We review the various aspects of luminescence thermometry, including different types of luminescence measurements and the numerous materials used as luminescence sensors. We end the article by highlighting other physical quantities that can be measured by luminescence (e.g., pressure, electric field strength, magnetic field strength), and provide a brief overview of applications of luminescence thermometry in biomedicine.

Pacemaker implantation-related infective endocarditis is an uncommon, but serious complication. The estimated incidence of infection after permanent endocardial pacemaker implantation varies from 0.2% to 3%. Infective endocarditis shows a mortality rate of 30% to 35%. Conservative medical treatment without removing the pacemaker is prone to failure, and reinfection rates of 51% to 77% have been noted in patients whose pacemaker systems become infected. Therefore, diagnosis of the primary infection is particularly important because surgical removal of the pacemaker system is usually required for its treatment. We describe here an 80-year-old woman who had drug fever that could have been misdiagnosed as infective endocarditis after pacemaker implantation.

Mucormycosis is a rare invasive disease with high mortality rates caused by fungi of the zygomycetes class and Mucorales order. Mucormycosis is manifested by a variety of clinical presentations according to the involved site and occurs in immunocompromised conditions such as diabetes mellitus and other conditions. Rhino-orbital-cerebral and pulmonary infection is known as commonly involved areas. Primary gastrointestinal mucormycosis is a very rare and life-threatening invasive fungal infection. Gastrointestinal mucormycosis may occur in any alimentary tract, with the stomach being most involved part. The early diagnosis and appropriate treatment including surgical debridement of involved tissues and antifungal agents is needed to improve survival rates. We report a case of gastric mucormycosis successfully treated with radical debridement and antifungal agents in a 45-year-old man with diabetic ketoacidosis.