Emergency Patient Triage Improvement through a Retrieval-Augmented Generation Enhanced Large-Scale Language Model.

Objectives: Emergency medical triage is crucial for prioritizing patient care in emergency situations, yet its effectiveness can vary significantly based on the experience and training of the personnel involved. This study aims to evaluate the efficacy of integrating Retrieval Augmented Generation (RAG) with Large Language Models (LLMs), specifically OpenAI's GPT models, to standardize triage procedures and reduce variability in emergency care.Methods: We created 100 simulated triage scenarios based on modified cases from the Japanese National Examination for Emergency Medical Technicians. These scenarios were processed by the RAG-enhanced LLMs, and the models were given patient vital signs, symptoms, and observations from emergency medical services (EMS) teams as inputs. The primary outcome was the accuracy of triage classifications, which was used to compare the performance of the RAG-enhanced LLMs to that of emergency medical technicians and emergency physicians. Secondary outcomes included the rates of under-triage and over-triage.Results: The Generative Pre-trained Transformer 3.5 (GPT-3.5) with RAG model achieved a correct triage rate of 70%, significantly outperforming Emergency Medical Technicians (EMTs) with 35% and 38% correct rates, and emergency physicians with 50% and 47% correct rates (p < 0.05). Additionally, this model demonstrated a substantial reduction in under-triage rates to 8%, compared to 33% for GPT-3.5 without RAG, and 39% for GPT-4 without RAG.Conclusions: The integration of RAG with LLMs shows promise in improving the accuracy and consistency of medical assessments in emergency settings. Further validation in diverse medical settings with broader datasets is necessary to confirm the effectiveness and adaptability of these technologies in live environments.

Automated detection and classification of the rotator cuff tear on plain shoulder radiograph using deep learning.

BACKGROUND: The diagnosis of rotator cuff tears (RCTs) using radiographs alone is clinically challenging; thus, the utility of deep learning algorithms based on convolutional neural networks has been remarkable in the field of medical imaging recognition. We aimed to evaluate the diagnostic performance of artificial intelligence (a deep learning algorithm; a convolutional neural network) to detect and classify RCTs using shoulder radiographs, and compare its diagnostic performance with that of orthopedic surgeons. METHODS: A total of 1169 plain shoulder anteroposterior radiographs (1 image per shoulder) were included in the total dataset and divided into four groups: intact, small, medium, and large to massive tear groups. The sensitivity, specificity, positive predictive value, negative predictive value, accuracy, and area under the receiver operating curve were measured for the detection of RCTs through binary classification. The average accuracy, recall, precision, and F1-score were divided into four groups by cuff tear size for multiclass classification. RESULTS: The convolutional neural network demonstrated a high performance, with 92% sensitivity, 69% specificity, 86% accuracy, and an area under the receiver operating curve of 0.88 for the detection of RCTs. The average accuracy, recall, precision, and F1-score of the convolutional neural network for classification were 60%, 0.42, 0.49, and 0.45, respectively. The accuracy of the convolutional neural network for the detection and classification of RCTs was significantly better than that of orthopedic surgeons. CONCLUSION: The convolutional neural network demonstrated the diagnostic ability to detect and classify RCTs using plain shoulder radiographs, and the diagnostic performance exhibited equal to superior accuracy when compared with those of shoulder experts.

Mapping the Japanese orthopedic association national registry (JOANR) to the international classification of health interventions (ICHI).

BACKGROUND: The Japanese Orthopedic Association launched the Japanese Orthopedic Association National Registry (JOANR), Japan's first large-scale nationwide musculoskeletal disease registry, in 2020. The World Health Organization released the International Classification of Health Interventions (ICHI) Beta-3 version in the same year. This concurrence served as an impetus to examine the relationship between domestic and international classification for orthopedic interventions. Our objective was to evaluate the possibility of utilizing JOANR for international comparison and the potential usage of ICHI in the domestic medical fee reimbursement system. This study is a novel attempt at mapping a domestic orthopedic scheme to the ICHI. METHODS: We mapped 149 codes out of 581 orthopedic surgical codes, on JOANR's registration form, to the ICHI, and then classified the nature of JOANR codes' relationship, to both ICHI single stem codes and stem codes accompanied by other additional stem codes, extension codes, and International Classification of Diseases for Mortality and Morbidity Statistics (ICD) codes, into five categories: Equivalent (exact match), Narrower (compared to ICHI; can be smoothly incorporated into ICHI), Broader (compared to ICHI), Slipped (combination of both Narrower and Broader), and None (no appropriate code). Finally, debatable issues that arose during the mapping operation were noted. RESULTS: The domestic codes' relationship to ICHI single stem code by category were Equivalent: 27 (18.1%) and Narrower: 65 (43.6%), respectively. Further, the rate of Equivalent rose to 120 (80.5%) on adding other stem codes, extension codes, and ICD codes. Additionally, certain domestic titles, which were unsuitable for classification as they included diagnostic information, and arthroscopic surgeries without corresponding ICHI codes, were recoded. CONCLUSIONS: JOANR can be converted to an international comparison standard via ICHI to a certain extent, and ICHI accompanied by ICD codes has potential for deployment in the domestic medical fee reimbursement system.

Quantitative evaluation of the lumbar ligamentum flavum using MRI T2-mapping: Efficacy of its clinical application in patients with lumbar spinal stenosis.

OBEJECTIVE: To perform a magnetic resonance imaging T2-mapping of the ligamentum flavum in healthy individuals and patients with lumbar spinal stenosis scheduled for surgery and compare the T2 relaxation times. SUBJECTS AND METHODS: The T2 relaxation time of the ligamentum flavum was compared among 3 groups, healthy young individuals (H group (age< 50)), healthy middle-aged and older individuals (H group (age≥50)), and patients with lumbar spinal stenosis (L group). Additionally, the thickness of the ligament was measured in the axial image plane, and the occupied area ratio of each fiber was measured by staining the surgically obtained ligament, and each was correlated with the T2 relaxation time. We also evaluated the adhesion of the ligamentum flavum with the dura mater during the surgery. RESULTS: The T2 relaxation times were significantly prolonged in H group (age ≥50) and L group (P < 0.001) compared to H group (age<50). The relationship between collagen fiber and T2 relaxation times was significantly positive (r = 0.720, P < 0.001). Moreover, the relaxation times were significantly prolonged in those with adhesion of the ligamentum flavum with the dura mater (P < 0.05). The cut-off for the relaxation time was 50 ms (sensitivity: 62.50%, false positive rate: 10.8%). CONCLUSION: Healthy middle-aged and older individuals and patients with lumbar spinal stenosis and adhesion of the ligamentum flavum with the dura mater have prolonged T2 relaxation times. Hence, the adhesion between the ligamentum flavum and dura mater should be considered in cases with a relaxation time ≥50 ms.

Deep learning-based prediction model for postoperative complications of cervical posterior longitudinal ligament ossification.

PURPOSE: Postoperative complication prediction helps surgeons to inform and manage patient expectations. Deep learning, a model that finds patterns in large samples of data, outperform traditional statistical methods in making predictions. This study aimed to create a deep learning-based model (DLM) to predict postoperative complications in patients with cervical ossification of the posterior longitudinal ligament (OPLL). METHODS: This prospective multicenter study was conducted by the 28 institutions, and 478 patients were included in the analysis. Deep learning was used to create two predictive models of the overall postoperative complications and neurological complications, one of the major complications. These models were constructed by learning the patient's preoperative background, clinical symptoms, surgical procedures, and imaging findings. These logistic regression models were also created, and these accuracies were compared with those of the DLM. RESULTS: Overall complications were observed in 127 cases (26.6%). The accuracy of the DLM was 74.6 ± 3.7% for predicting the overall occurrence of complications, which was comparable to that of the logistic regression (74.1%). Neurological complications were observed in 48 cases (10.0%), and the accuracy of the DLM was 91.7 ± 3.5%, which was higher than that of the logistic regression (90.1%). CONCLUSION: A new algorithm using deep learning was able to predict complications after cervical OPLL surgery. This model was well calibrated, with prediction accuracy comparable to that of regression models. The accuracy remained high even for predicting only neurological complications, for which the case number is limited compared to conventional statistical methods.

A 2-year longitudinal study of skeletal muscle mass in women over 40 years of age with degenerative lumbar scoliosis.

PURPOSE: We investigated changes in skeletal muscle mass and bone mineral density in degenerative lumbar scoliosis (DLS) patients during a 2-year follow-up following diagnosis. METHOD: This study included 418 Japanese women, identifying 50 patients for the DLS group (mean age 76.4 years) and 368 patients for the control group (mean age 73.4 years). Whole-body skeletal muscle mass was measured using a Bioelectrical Impedance Analyzer. Bone mineral density (BMD) was measured using DXA. Skin autofluorescence (SAF), a marker of advanced glycation end products in the skin, was measured using a spectroscope. Spinal alignment, skeletal muscle mass, BMD, grip strength, and SAF were examined and the amount of change 1 and 2 years from the initial examination for each item was compared between groups. RESULTS: Height, body fat mass, grip strength, upper limb muscle mass, and trunk muscle mass in the DLS group were significantly lower, and lumbar spine BMD was significantly greater compared to controls at the first visit (p < 0.05). There was no significant difference in spinal alignment in the DLS group after 2 years compared with baseline. Trunk muscle mass also decreased significantly more in the DLS group (-2.7%) than in the control group (-1.1%) over the 2-year follow-up (p < 0.05). DISCUSSION: In this study, trunk muscle mass in the DLS group decreased about 2.4 times more in 2 years compared with the control group (p < 0.05). It may be possible to clarify the mechanism of kyphoscoliosis progression in the future with large-scale longitudinal studies.

MRI T2-mapping of lumbar facet joints is effective for quantitative evaluation of lumbar instability in patients with degenerative lumbar disorders.

PURPOSES: To analyze T2 relaxation times of the facet joint by MRI T2-mapping in patients with degenerative lumbar disorders (DLD), and to determine the correlation with lumbar instability in radiographs. METHODS: We conducted a T2-mapping of the lumbar facet joint using a 1.5 T MRI system. We classified patients with degenerative lumbar disorders scheduled to undergo decompression surgery into groups with stability and instability using radiographs, and compared the T2 relaxation times of the lumbar facet. Lumbar instability was defined as the presence of anterior translation ratio > 5% or disk range of motion (ROM) > 5° in the sagittal plane of SLFE radiographs. RESULTS: Inclusion criteria were met by 22 patients (45 levels, mean age 64.3 years). Facet effusions had high sensitivity (90%) but had low specificity (28%) for diagnosis of lumbar instability. Mean T2 relaxation times of right and left facet joints are significantly longer (98.4 ms) in the instability group than they are (87.6 ms) in the stability group (p < 0.001). Anterior translation ratio was positively correlated with mean T2 relaxation times of facet joint (R2 = 0.493, p < 0.05). From a ROC analysis, the cutoff value of T2 relaxation times for lumbar instability was 98.65 ms (sensitivity 60.0%, specificity 95.7%, AUC 0.763). CONCLUSIONS: The T2 relaxation times were positively correlated with lumbar instability. This new quantitative evaluation of lumbar facet joint using MRI T2-mapping might be useful to determine lumbar instability.

Decreased muscle mass and strength affected spinal sagittal malalignment.

PURPOSE: Correction surgeries for spinal malalignment showed good clinical outcomes; however, there were concerns including increased invasiveness, complications, and impact on medico-economics. Ideally, an early intervention is needed. To better understand the patho-mechanism and natural course of spinal alignment, the effect of factors such as muscle mass and strength on spinal sagittal imbalance were determined in a multicenter cross-sectional study. METHODS: After excluding metal implant recipients, 1823 of 2551 patients (mean age: 69.2 ± 13.8 years; men 768, women 1055) were enrolled. Age, sex, past medical history (Charlson comorbidity index), body mass index (BMI), grip strength (GS), and trunk muscle mass (TM) were reviewed. Spinal sagittal imbalance was determined by the SRS-Schwab classification. Multiple comparison analysis among four groups (Normal, Mild, Moderate, Severe) and multinomial logistic regression analysis were performed. RESULTS: On multiple comparison analysis, with progressing spinal malalignment, age in both sexes tended to be higher; further, TM in women and GS in both sexes tended to be low. On multinomial logistic regression analysis, age and BMI were positively associated with spinal sagittal malalignment in Mild, Moderate, and Severe groups. TM in Moderate and Severe groups and GS in the Moderate group were negatively associated with spinal sagittal malalignment. CONCLUSION: Aging, obesity, low TM, and low GS are potential risk factors for spinal sagittal malalignment. Especially, low TM and low GS are potentially associated with more progressed spinal sagittal malalignment. Thus, early intervention for muscles, such as exercise therapy, is needed, while the spinal sagittal alignment is normal or mildly affected.

Automated detection of anterior cruciate ligament tears using a deep convolutional neural network.

BACKGROUND: The development of computer-assisted technologies to diagnose anterior cruciate ligament (ACL) injury by analyzing knee magnetic resonance images (MRI) would be beneficial, and convolutional neural network (CNN)-based deep learning approaches may offer a solution. This study aimed to evaluate the accuracy of a CNN system in diagnosing ACL ruptures by a single slice from a knee MRI and to compare the results with that of experienced human readers. METHODS: One hundred sagittal MR images from patients with and without ACL injuries, confirmed by arthroscopy, were cropped and used for the CNN training. The final decision by the CNN for intact or torn ACL was based on the probability of ACL tear on a single MRI slice. Twelve board-certified physicians reviewed the same images used by CNN. RESULTS: The sensitivity, specificity, accuracy, positive predictive value and negative predictive value of the CNN classification was 91.0%, 86.0%, 88.5%, 87.0%, and 91.0%, respectively. The overall values of the physicians' readings were similar, but the specificity was lower than the CNN classification for some of the physicians, thus resulting in lower accuracy for the human readers. CONCLUSIONS: The trained CNN automatically detected the ACL tears with acceptable accuracy comparable to that of human readers.

Analgesic effects and arthritic changes following intra-articular injection of diclofenac etalhyaluronate in a rat knee osteoarthritis model.

BACKGROUND: Diclofenac etalhyaluronate (DF-HA) is a recently developed analgesic conjugate of diclofenac and hyaluronic acid that has analgesic and anti-inflammatory effects on acute arthritis. In this study, we investigated its analgesic effect on osteoarthritis, using a rat model of monoiodoacetate (MIA). METHODS: We injected MIA into the right knees of eight 6-weeks-old male Sprague-Dawley rats. Four weeks later, rats were randomly injected with DF-HA or vehicle into the right knee. Seven weeks after the MIA injection, fluorogold (FG) and sterile saline were injected into the right knees of all the rats. We assessed hyperalgesia with weekly von Frey tests for 8 weeks after MIA administration. We took the right knee computed tomography (CT) as radiographical evaluation every 2 weeks. All rats were sacrificed 8 weeks after administration of MIA for histological evaluation of the right knee and immunohistochemical evaluation of the DRG and spinal cord. We also evaluated the number of FG-labeled calcitonin gene-related peptide (CGRP)-immunoreactive(ir) neurons in the dorsal root ganglion (DRG) and ionized calcium-binding adapter molecule 1 (Iba1)-ir microglia in the spinal cord. RESULTS: Administration of DF-HA significantly improved pain sensitivity and reduced CGRP and Iba1 expression in the DRG and spinal cord, respectively. However, computed tomography and histological evaluation of the right knee showed similar levels of joint deformity, despite DF-HA administration. CONCLUSION: DF-HA exerted analgesic effects on osteoarthritic pain, but did not affect joint deformity.

Assessment of locomotive syndrome in patients with visceral cancer, the comparison with non-cancer patients using propensity score matching.

BACKGROUND: When treating cancer patients, the progression of symptoms is accompanied by the deterioration of systemic conditions and motor function. From a risk-benefit perspective, a certain level of physical function must be maintained to continue cancer treatment. Recently, outpatient cancer treatment has become more common. Motor function is important to determine the feasibility of continuing cancer treatment. The study aimed to evaluate the motor function of patients with visceral cancer using locomo tests established by Japanese Orthopaedic Association. METHODS: Locomo tests were performed, and the results were compared with data from non-cancer individuals. Background data were matched by propensity score matching. Data from 53 cancer patients (group C) were compared with that of 75 non-cancer patients (group N). RESULTS: The average score in the two-step test of group C was lower than that of group N (1.27: 1.37, p = 0.004). The average function in the stand-up test of group C was worse than that of group N (p = 0.001). The average score in the 25-question geriatric locomotive function scale (GLFS) of group C was significantly higher than that of group N (19.92: 5.29, SE 2.21, p < 0.001). Higher 25-question GLFS scores indicate reduced mobility. The proportion of the locomo stage 2 in group C was significantly higher than in group N (51%: 13%, p < 0.001). The results of the two field tests revealed a clinically minimal difference between the two groups, but a statistically significant difference. Locomo tests may be detect potential motor dysfunction in outpatient cancer patients with apparently maintained motor function. CONCLUSIONS: Even in cancer patients who attend outpatient clinics, their motor functions could be potentially impaired. Therapeutic interventions to maintain and enhance motor function for cancer patients could be useful for continuing cancer treatment, and furthermore, improving prognosis.

The characteristics of the young patients with cervical ossification of the posterior longitudinal ligament of the spine: A multicenter cross-sectional study.

BACKGROUND: Ossification of the posterior longitudinal ligament of the spine (OPLL) is characterized by heterotopic bone formation in the posterior longitudinal ligament of the spine. Although the patients with OPLL are more common in the 60s and 70s, we know that there are markedly young patients (e.g., early 40s). However, to the best of our knowledge, there is few reports characterize young patients with cervical OPLL in terms of the imaging features, subjective symptoms, and ADL problems. METHODS: This is the multicenter cross-sectional study. Two hundred and thirty-seven Japanese symptomatic patients with cervical OPLL confirmed by standard X-rays collected from 16 institutions belonging to the Japanese Multicenter Research Organization for Ossification of the Spinal Ligament formed by the Japanese Ministry of Health, Labor and Welfare were recruited. Whole spine CT data as well as demographic data such as age, gender, patients-based evaluations, and the 36-item Short Form Health Survey (SF-36) were evaluated. RESULTS: Young group (≦ 45 years old) consisted of 23 patients (8 females and 15 males), accounting for 9.7% of the total. Their characteristics were high body mass index (BMI), significant involvement of trauma in the onset and deterioration of symptoms, and the predominance of thoracic OPLL. The patient-based evaluations did not show a significant difference between the young and non-young groups, or between the genders in the young group except for bodily pain (BP) of SF-36. Female patients in young group had significantly lower BP score of SF-36 than that of male in young group. CONCLUSIONS: Characteristics of young patients with cervical OPLL were high BMI, significant involvement of trauma in the onset and deterioration of symptoms, lower BP score of SF-36 in female, and the predominance of thoracic OPLL.

Detecting Distal Radial Fractures from Wrist Radiographs Using a Deep Convolutional Neural Network with an Accuracy Comparable to Hand Orthopedic Surgeons.

In recent years, fracture image diagnosis using a convolutional neural network (CNN) has been reported. The purpose of the present study was to evaluate the ability of CNN to diagnose distal radius fractures (DRFs) using frontal and lateral wrist radiographs. We included 503 cases of DRF diagnosed by plain radiographs and 289 cases without fracture. We implemented the CNN model using Keras and Tensorflow. Frontal and lateral views of wrist radiographs were manually cropped and trained separately. Fine-tuning was performed using EfficientNets. The diagnostic ability of CNN was evaluated using 150 images with and without fractures from anteroposterior and lateral radiographs. The CNN model diagnosed DRF based on three views: frontal view, lateral view, and both frontal and lateral view. We determined the sensitivity, specificity, and accuracy of the CNN model, plotted a receiver operating characteristic (ROC) curve, and calculated the area under the ROC curve (AUC). We further compared performances between the CNN and three hand orthopedic surgeons. EfficientNet-B2 in the frontal view and EfficientNet-B4 in the lateral view showed highest accuracy on the validation dataset, and these models were used for combined views. The accuracy, sensitivity, and specificity of the CNN based on both anteroposterior and lateral radiographs were 99.3, 98.7, and 100, respectively. The accuracy of the CNN was equal to or better than that of three orthopedic surgeons. The AUC of the CNN on the combined views was 0.993. The CNN model exhibited high accuracy in the diagnosis of distal radius fracture with a plain radiograph.

Machine learning-based image analysis for accelerating the diagnosis of complicated preneoplastic and neoplastic ductal lesions in breast biopsy tissues.

PURPOSE: Diagnosis of breast preneoplastic and neoplastic lesions is difficult due to their similar morphology in breast biopsy specimens. To diagnose these lesions, pathologists perform immunohistochemical analysis and consult with expert breast pathologists. These additional examinations are time-consuming and expensive. Artificial intelligence (AI)-based image analysis has recently improved, and may help in ordinal pathological diagnosis. Here, we showed the significance of machine learning-based image analysis of breast preneoplastic and neoplastic lesions for facilitating high-throughput diagnosis. METHODS: Images were obtained from normal mammary glands, hyperplastic lesions, preneoplastic lesions and neoplastic lesions, such as usual ductal hyperplasia (UDH), columnar cell lesion (CCL), ductal carcinoma in situ (DCIS), and DCIS with comedo necrosis (comedo DCIS) in breast biopsy specimens. The original enhanced convoluted neural network (CNN) system was used for analyzing the pathological images. RESULTS: The AI-based image analysis provided the following area under the curve values (AUC): normal lesion versus DCIS, 0.9902; DCIS versus comedo DCIS, 0.9942; normal lesion versus CCL, 0.9786; and UDH versus DCIS, 1.000. Multiple comparison analysis showed precision and recall scores similar to those of single comparison analysis. Based on the gradient-weighted class activation mapping (Grad-CAM) used to visualize the important regions reflecting the result of CNN analysis, the ratio of stromal tissue in the whole weighted area was significantly higher in UDH and CCL than that in DCIS. CONCLUSIONS: These analyses may provide a more accurate and rapid pathological diagnosis of patients. Moreover, Grad-CAM identifies uncharted important histological characteristics for newer pathological findings and targets of research for understanding diseases.

Multi-shot acquisitions for stimulus-evoked spinal cord BOLD fMRI.

PURPOSE: To demonstrate the feasibility of 3D multi-shot magnetic resonance imaging acquisitions for stimulus-evoked blood oxygenation level dependent (BOLD) functional magnetic resonance imaging (fMRI) in the human spinal cord in vivo. METHODS: Two fMRI studies were performed at 3T. The first study was a hypercapnic gas challenge where data were acquired from healthy volunteers using a multi-shot 3D fast field echo (FFE) sequence as well as single-shot multi-slice echo-planar imaging (EPI). In the second study, another cohort of healthy volunteers performed an upper extremity motor task while fMRI data were acquired using a 3D multi-shot acquisition. RESULTS: Both 2D-EPI and 3D-FFE were shown to be sensitive to BOLD signal changes in the cervical spinal cord, and had comparable contrast-to-noise ratios in gray matter. FFE exhibited much less signal drop-out and weaker geometric distortions compared to EPI. In the motor paradigm study, the mean number of active voxels was highest in the ventral gray matter horns ipsilateral to the side of the task and at the spinal level associated with innervation of finger extensors. CONCLUSIONS: Highly multi-shot acquisition sequences such as 3D-FFE are well suited for stimulus-evoked spinal cord BOLD fMRI.

Minimally displaced unilateral facet fracture of cervical spine can lead to spinal cord injury: a report of two cases.

BACKGROUND: According to most of the commonly used classification systems for subaxial spine injuries, unilateral and minimally displaced facet fractures without any sign of a spinal cord injury would be directed to non-operative management. However, the failure rate of non-operative treatment varies from 20 to 80%, and no consensus exists with regard to predictors of failure after non-operative management. CASE PRESENTATION: Case 1 is a patient with a unilateral facet fracture. The patient had only numbness in the right C6 dermatome but failed non-operative treatment, which resulted in severe spinal cord injury. Case 2 is a patient who had a similar injury pattern as case 1 but presented with immediate instability and underwent fusion surgery. Both patients had a minimally displaced unilateral facet fracture accompanied by disc injury and blunt vertebral artery injury, which are possible signs indicating significant instability. CONCLUSIONS: This is the first report of an isolated unilateral facet fracture that resulted in catastrophic spinal cord injury. These two cases illustrate that an isolated minimally displaced unilateral facet fracture with disc injury and vertebral artery injury were associated with significant instability that can lead to spinal cord injury.

Characteristics of relief and residual low back pain after discectomy in patients with lumbar disc herniation: analysis using a detailed visual analog scale.

BACKGROUND: Several authors have reported favorable results in low back pain (LBP) for patients with lumbar disc herniation (LDH) treated with discectomy. However, detailed changes over time in the characteristics and location of LBP before and after discectomy for LDH remain unclear. To clarify these points, we conducted an observational study to determine the detailed characteristics and location of LBP before and after discectomy for LDH, using a detailed visual analog scale (VAS) bilaterally. METHODS: We included 65 patients with LDH treated by discectomy in this study. A detailed VAS for LBP was administered with the patient under 3 different conditions: in motion, standing, and sitting. Bilateral VAS was also administered (affected versus opposite side) for LBP, lower extremity pain (LEP), and lower extremity numbness (LEN). The Oswestry Disability Index (ODI) was used to quantify clinical status. Changes over time in these VAS and ODI were investigated. Pfirrmann grading and Modic change as seen by magnetic resonance imaging (MRI) were reviewed before and 1 year after discectomy to determine disc and endplate condition. RESULTS: Before surgery, LBP on the affected side while the patients were in motion was significantly higher than LBP while they were sitting (p = 0.025). This increased LBP on the affected side in motion was improved significantly after discectomy (p < 0.001). By contrast, the residual LBP while sitting at 1 year after surgery was significantly higher than the LBP while they were in motion or standing (p = 0.015). At 1 year following discectomy, residual LBP while sitting was significantly greater in cases showing changes in Pfirrmann grade (p = 0.002) or Modic type (p = 0.025). CONCLUSIONS: Improvement of LBP on the affected side while the patient is in motion suggests that radicular LBP is improved following discectomy by nerve root decompression. Furthermore, residual LBP may reflect increased load and pressure on the disc and endplate in the sitting position.

Does the increased motion probing gradient directional diffusion tensor imaging of lumbar nerves using multi-band SENSE improve the visualization and accuracy of FA values?

PURPOSE: Diffusion tensor imaging (DTI) is useful to evaluate lumbar nerves visually and quantitatively. Multi-band sensitivity encoding (MB-SENSE) is a technique to reduce the scan time. This study aimed to investigate if super-multi-gradient DTI with multi-band sensitivity encoding (MB-SENSE) is better in evaluating lumbar nerves than the conventional method. METHODS: The participants were 12 healthy volunteers (mean age 33.6 years). In all subjects, DTI was performed using echo planar imaging with different motion probing gradient (MPG) directions (15 without MB, and 15, 32, 64, and 128 with MB) and the lumbar nerve roots were visualized with tractography. In the five groups, we evaluated the resultant DTI both visually and quantitatively. For visual measures, we counted the number of fluffs and disruptions of the nerve fibers. For quantitative measures, the fractional anisotropy (FA) and standard deviation of the fractional anisotropy (FA-SD) values at two regions (proximal and distal) of the lumbar nerve roots were quantified and compared. RESULTS: Among the five groups, the number of fluffs decreased as the number of MPG directions increased. However, the number of disruptions showed no significant differences. The FA-SD values decreased as the number of MPG directions increased, indicating that the signal variation was reduced with multi-gradient directional DTI. CONCLUSION: High-resolution multi-directional DTI with MB-SENSE may be useful to visualize nerve entrapments and may allow for more accurate DTI parameter quantification with opportunities for clinical diagnostic applications.

Automated classification of hip fractures using deep convolutional neural networks with orthopedic surgeon-level accuracy: ensemble decision-making with antero-posterior and lateral radiographs.

Background and purpose - Deep-learning approaches based on convolutional neural networks (CNNs) are gaining interest in the medical imaging field. We evaluated the diagnostic performance of a CNN to discriminate femoral neck fractures, trochanteric fractures, and non-fracture using antero-posterior (AP) and lateral hip radiographs. Patients and methods - 1,703 plain hip AP radiographs and 1,220 plain hip lateral radiographs were included in the total dataset. 150 images each of the AP and lateral views were separated out and the remainder of the dataset was used for training. The CNN made the diagnosis based on: (1) AP radiographs alone, (2) lateral radiographs alone, or (3) both AP and lateral radiographs combined. The diagnostic performance of the CNN was measured by the accuracy, recall, precision, and F1 score. We further compared the CNN's performance with that of orthopedic surgeons. Results - The average accuracy, recall, precision, and F1 score of the CNN based on both anteroposterior and lateral radiographs were 0.98, 0.98, 0.98, and 0.98, respectively. The accuracy of the CNN was comparable to, or statistically significantly better than, that of the orthopedic surgeons regardless of radiographic view used. In the CNN model, the accuracy of the diagnosis based on both views was significantly better than the lateral view alone and tended to be better than the AP view alone. Interpretation - The CNN exhibited comparable or superior performance to that of orthopedic surgeons to discriminate femoral neck fractures, trochanteric fractures, and non-fracture using both AP and lateral hip radiographs.

Multiple sclerosis lesions affect intrinsic functional connectivity of the spinal cord.

Patients with multiple sclerosis present with focal lesions throughout the spinal cord. There is a clinical need for non-invasive measurements of spinal cord activity and functional organization in multiple sclerosis, given the cord's critical role in the disease. Recent reports of spontaneous blood oxygenation level-dependent fluctuations in the spinal cord using functional MRI suggest that, like the brain, cord activity at rest is organized into distinct, synchronized functional networks among grey matter regions, likely related to motor and sensory systems. Previous studies looking at stimulus-evoked activity in the spinal cord of patients with multiple sclerosis have demonstrated increased levels of activation as well as a more bilateral distribution of activity compared to controls. Functional connectivity studies of brain networks in multiple sclerosis have revealed widespread alterations, which may take on a dynamic trajectory over the course of the disease, with compensatory increases in connectivity followed by decreases associated with structural damage. We build upon this literature by examining functional connectivity in the spinal cord of patients with multiple sclerosis. Using ultra-high field 7 T imaging along with processing strategies for robust spinal cord functional MRI and lesion identification, the present study assessed functional connectivity within cervical cord grey matter of patients with relapsing-remitting multiple sclerosis (n = 22) compared to a large sample of healthy controls (n = 56). Patient anatomical images were rated for lesions by three independent raters, with consensus ratings revealing 19 of 22 patients presented with lesions somewhere in the imaged volume. Linear mixed models were used to assess effects of lesion location on functional connectivity. Analysis in control subjects demonstrated a robust pattern of connectivity among ventral grey matter regions as well as a distinct network among dorsal regions. A gender effect was also observed in controls whereby females demonstrated higher ventral network connectivity. Wilcoxon rank-sum tests detected no differences in average connectivity or power of low frequency fluctuations in patients compared to controls. The presence of lesions was, however, associated with local alterations in connectivity with differential effects depending on columnar location. The patient results suggest that spinal cord functional networks are generally intact in relapsing-remitting multiple sclerosis but that lesions are associated with focal abnormalities in intrinsic connectivity. These findings are discussed in light of the current literature on spinal cord functional MRI and the potential neurological underpinnings.