Autonomous Chest Radiograph Reporting Using AI: Estimation of Clinical Impact.

Background Automated interpretation of normal chest radiographs could alleviate the workload of radiologists. However, the performance of such an artificial intelligence (AI) tool compared with clinical radiology reports has not been established. Purpose To perform an external evaluation of a commercially available AI tool for (a) the number of chest radiographs autonomously reported, (b) the sensitivity for AI detection of abnormal chest radiographs, and (c) the performance of AI compared with that of the clinical radiology reports. Materials and Methods In this retrospective study, consecutive posteroanterior chest radiographs from adult patients in four hospitals in the capital region of Denmark were obtained in January 2020, including images from emergency department patients, in-hospital patients, and outpatients. Three thoracic radiologists labeled chest radiographs in a reference standard based on chest radiograph findings into the following categories: critical, other remarkable, unremarkable, or normal (no abnormalities). AI classified chest radiographs as high confidence normal (normal) or not high confidence normal (abnormal). Results A total of 1529 patients were included for analysis (median age, 69 years [IQR, 55-69 years]; 776 women), with 1100 (72%) classified by the reference standard as having abnormal radiographs, 617 (40%) as having critical abnormal radiographs, and 429 (28%) as having normal radiographs. For comparison, clinical radiology reports were classified based on the text and insufficient reports excluded (n = 22). The sensitivity of AI was 99.1% (95% CI: 98.3, 99.6; 1090 of 1100 patients) for abnormal radiographs and 99.8% (95% CI: 99.1, 99.9; 616 of 617 patients) for critical radiographs. Corresponding sensitivities for radiologist reports were 72.3% (95% CI: 69.5, 74.9; 779 of 1078 patients) and 93.5% (95% CI: 91.2, 95.3; 558 of 597 patients), respectively. Specificity of AI, and hence the potential autonomous reporting rate, was 28.0% of all normal posteroanterior chest radiographs (95% CI: 23.8, 32.5; 120 of 429 patients), or 7.8% (120 of 1529 patients) of all posteroanterior chest radiographs. Conclusion Of all normal posteroanterior chest radiographs, 28% were autonomously reported by AI with a sensitivity for any abnormalities higher than 99%. This corresponded to 7.8% of the entire posteroanterior chest radiograph production. © RSNA, 2023 Supplemental material is available for this article. See also the editorial by Park in this issue.

Magnetic Resonance T2 * Is Increased in Patients With Early-Stage Achilles and Patellar Tendinopathy.

BACKGROUND: T2 * mapping has proven useful in tendon research and may have the ability to detect subtle changes at an early stage of tendinopathy. PURPOSE: To investigate the difference in T2 * between patients with early tendinopathy and healthy controls, and to investigate the relationship between T2 * and clinical outcomes, tendon size, and mechanical properties. STUDY TYPE: Prospective cross-sectional. SUBJECTS: Sixty-five patients with early tendinopathy and 25 healthy controls. FIELD STRENGTH/SEQUENCE: Three Tesla, ultrashort time to echo magnetic resonance imaging. ASSESSMENT: Tendon T2 * was quantified using a monoexponential fitting algorithm. Clinical symptoms were evaluated using the Victorian Institute of Sports Assessment-Achilles/Patella (VISA-A/VISA-P). In vivo mechanical properties were measured using an ultrasound-based method that determined force and deformation simultaneously in tendons of patellar tendinopathy patients. STATISTICAL TESTS: A generalized linear model adjusted for age was applied to investigate the difference between patients and controls. In the two patient groups, linear regressions were applied to investigate the association between T2 * and tendon size, clinical outcomes, and biomechanical properties. RESULTS: There was a significant difference in T2 * between patients and healthy controls (204.8 [95% CI: 44.5-365.0] µsec, P < 0.05). There was a positive correlation between tendon size and T2 * for both Achilles (r = 0.72; P < 0.05) and patellar tendons (r = 0.53; P < 0.05). There was no significant correlation between VISA-A and T2 * (r = -0.2; P = 0.17) or VISA-P and T2 * (r = -0.5; P = 0.0504). Lastly, there was a negative correlation between modulus and T2 * (r = -0.51; P < 0.05). DATA CONCLUSIONS: T2 * mapping can detect subtle structural changes that translate to altered mechanical properties in early-phase tendinopathy. However, T2 * did not correlate with clinical scores in patients with early-phase Achilles and patellar tendinopathy. Thus, T2 * mapping may serve as a tool for early detection of structural changes in tendinopathy but does not necessarily describe the clinical severity of disease. LEVEL OF EVIDENCE: 1 TECHNICAL EFFICACY STAGE: 2.

Early development of tendinopathy in humans: Sequence of pathological changes in structure and tissue turnover signaling.

Overloading of tendon tissue with resulting chronic pain (tendinopathy) is a common disorder in occupational-, leisure- and sports-activity, but its pathogenesis remains poorly understood. To investigate the very early phase of tendinopathy, Achilles and patellar tendons were investigated in 200 physically active patients and 50 healthy control persons. Patients were divided into three groups: symptoms for 0-1 months (T1), 1-2 months (T2) or 2-3 months (T3). Tendinopathic Achilles tendon cross-sectional area determined by ultrasonography (US) was ~25% larger than in healthy control persons. Both Achilles and patellar anterior-posterior diameter were elevated in tendinopathy, and only later in Achilles was the width increased. Increased tendon size was accompanied by an increase in hypervascularization (US Doppler flow) without any change in mRNA for angiogenic factors. From patellar biopsies taken bilaterally, mRNA for most growth factors and tendon components remained unchanged (except for TGF-beta1 and substance-P) in early tendinopathy. Tendon stiffness remained unaltered over the first three months of tendinopathy and was similar to the asymptomatic contra-lateral tendon. In conclusion, this suggests that tendinopathy pathogenesis represents a disturbed tissue homeostasis with fluid accumulation. The disturbance is likely induced by repeated mechanical overloading rather than a partial rupture of the tendon.

UTE T2* mapping of tendinopathic patellar tendons: an MRI reproducibility study.

BACKGROUND: There is currently a lack of imaging modalities that can be used as a sensitive measure in tendinopathy. Recent findings suggest the applicability of ultra-short echo time (UTE) magnetic resonance imaging (MRI) T2* mapping in tendons, but the reproducibility remains unknown. PURPOSE: To evaluate test-retest reproducibility of UTE MRI T2* mapping of tendinopathic patellar tendons and to evaluate the intra- and inter-observer reproducibility of the measurement. MATERIAL AND METHODS: Fifteen patients with chronic patellar tendinopathy were evaluated with UTE MRI twice in a 3.0-T scanner on the same day. Manual segmentation of the patellar tendon was performed by two blinded investigators and automated T2*map reconstruction was performed in custom-made software. RESULTS: There was a significant and numerically small difference in test-retest T2* values (T2*meandiff = 0.06 ± 0.07 ms ≈ 3.7%; P = 0.006) with an ICC = 0.91 (95% confidence interval [CI] 0.58-0.98; typical error of 3.0%). The intra- and inter-observer reproducibility showed no significant bias (P = 0.493 and P = 0.052), and generally substantial reproducibility was demonstrated for T2* (intra-observer ICC = 0.99; 95% CI 0.98-1.00 and inter-observer ICC = 0.99; 95% CI 0.96-1.00, and typical error 1.3% and 1.3%, respectively). CONCLUSION: These data demonstrate a small bias between repeated measurements for UTE T2*, but with a very low associated mean difference (3.7%) between the two tests. The high ICC values and low typical error % demonstrate reproducibility of repeated T2*-mapping sessions. Further, the method showed substantial intra- and inter-observer reproducibility for T2* values proving feasibility for use of UTE T2* mapping in research and clinical practice.

Role of tissue perfusion, muscle strength recovery, and pain in rehabilitation after acute muscle strain injury: A randomized controlled trial comparing early and delayed rehabilitation.

Muscle strain injuries disrupt the muscle-tendon unit, early rehabilitation is associated with a faster return to sports (RTS), but the time course of tissue healing remains sparsely described. The purpose was to examine tissue regeneration and the effectiveness of early versus delayed rehabilitation onset on functional and structural recovery after strain injuries. A total of 50 recreational athletes with a severe acute strain injury in their thigh or calf muscles were randomized to early or delayed rehabilitation onset. Magnetic resonance imaging (MRI) was obtained initially, 3 and 6 months postinjury, and dynamic contrast-enhanced MRI (DCE-MRI) estimated tissue inflammation initially and after 6 months. Muscle strength was determined 5 weeks, 3 months, and 6 months postinjury, and a questionnaire determined soreness, pain, and confidence. DCE-MRI microvascular perfusion was higher in the injured compared to an uninjured muscle acutely (P < 0.01) and after 6 months (P < 0.01), for both groups (P > 0.05) and unrelated to RTS (P > 0.05). Total volume of the injured muscle decreased from the acute to the 3-month scan, and to the 6-month scan (P < 0.01) in both groups. Muscle strength was similar in both groups at any time. There was a nonsignificant trend (P ≤ 0.1) toward less pain and higher confidence with early rehabilitation. One reinjury was recorded. In conclusion, our data showed prolonged tissue repair with the initial response linked to muscle atrophy but did not explain why early rehabilitation onset accelerated recovery considering that structural and functional recovery was similar with early and delayed rehabilitation.

Point of no return for improvement of cartilage quality indicated by dGEMRIC before and after weight loss in patients with knee osteoarthritis: a cohort study.

Background It has been demonstrated that weight loss improves symptoms in obese subjects with knee osteoarthritis (KOA). A parallel change in cartilage morphology remains to be demonstrated. Purpose To demonstrate a parallel change in cartilage morphology. Material and Methods Obese patients with KOA were examined before and after weight loss over 16 weeks. Target knee joints were radiographically assessed by the Kellgren/Lawrence grading (KLG) system. Patients with KLG-1 and 2 changes in the lateral compartment were included. Delayed gadolinium-enhanced MRI of cartilage (dGEMRIC) was performed using intra-articular contrast. Results Nine patients with lateral KLG-1 and ten patients with lateral KLG-2 were studied. There were no group differences regarding the lateral compartment baseline dGEMRIC T1 values: median = 497 ms (KLG-1) and 533 ms (KLG-2) ( P = 0.12), or regarding reduction in body mass index (BMI) after 16 weeks: 12.8% versus 11.4% ( P = 0.74). In the KLG-1 group, several cases of increased dGEMRIC T1 values were seen and median value decreased significantly less than in KLG-2 group (15 ms versus 41 ms, P = 0.03) after weight loss. Conclusion Improvement of cartilage quality, assessed with dGEMRIC, after weight loss might be possible in early stage KOA (KLG-1), but not in later stage KOA (KLG-2). The results may suggest a point of no return for improvement of cartilage quality that should be tested in larger trials.

Navicular bone position determined by positional MRI: a reproducibility study.

OBJECTIVE: To examine intraobserver, interobserver and between-day reproducibility of positional MRI for evaluation of navicular bone height (NVH) and medial navicular position (MNP). MATERIALS AND METHODS: Positional MRI (pMRI) of the foot was performed on ten healthy participants (0.25 T G-scanner). Scanning was performed in supine and standing position, respectively. Two radiologists evaluated the images in a blinded manner. Reliability and agreement were assessed by calculation of intraclass correlation coefficient (ICC) and 95 % limits of agreement as a percentage of the mean (LOA%). RESULTS: Intraobserver and interobserver reliability was "substantial" in both supine and standing position (ICC 0.86-0.98) and showed good agreement (LOA% 4.9-14.7 %). Between-day reliability of navicular height and medial navicular position in standing position remained substantial (ICC 0.85-0.92) with adequate agreement (LOA% 8.3-19.8 %). In supine position between-day reliability was "moderate" for NVH (ICC 0.72) and "slight" for MNP (ICC 0.39). Agreement remained adequate between-days for MNP in supine position (LOA% 17.7 %), but it was less than adequate for NVH in supine position (LOA% 24.2 %). CONCLUSION: Navicular height and medial navicular position can be measured by pMRI in a very reproducible manner within and between observers. Increased measurement variation is observed between-days in supine position, which may be due to small positional differences or other unknown biomechanical factors.

Glucagon augments the secretion of FGF21 and GDF15 in MASLD by indirect mechanisms.

INTRODUCTION: Glucagon receptor agonism is currently explored for the treatment of obesity and metabolic dysfunction-associated steatotic liver disease (MASLD). The metabolic effects of glucagon receptor agonism may in part be mediated by increases in circulating levels of Fibroblast Growth Factor 21 (FGF21) and Growth Differentiation Factor 15 (GDF15). The effect of glucagon agonism on FGF21 and GDF15 levels remains uncertain, especially in the context of elevated insulin levels commonly observed in metabolic diseases. METHODS: We investigated the effect of a single bolus of glucagon and a continuous infusion of glucagon on plasma concentrations of FGF21 and GDF15 in conditions of endogenous low or high insulin levels. The studies included individuals with overweight with and without MASLD, healthy controls (CON) and individuals with type 1 diabetes (T1D). The direct effect of glucagon on FGF21 and GDF15 was evaluated using our in-house developed isolated perfused mouse liver model. RESULTS: FGF21 and GDF15 correlated with plasma levels of insulin, but not glucagon, and their secretion was highly increased in MASLD compared with CON and T1D. Furthermore, FGF21 levels in individuals with overweight with or without MASLD did not increase after glucagon stimulation when insulin levels were kept constant. FGF21 and GDF15 levels were unaffected by direct stimulation with glucagon in the isolated perfused mouse liver. CONCLUSION: The glucagon-induced secretion of FGF21 and GDF15 is augmented in MASLD and may depend on insulin. Thus, glucagon receptor agonism may augment its metabolic benefits in patients with MASLD through enhanced secretion of FGF21 and GDF15.

Development of a glucagon sensitivity test in humans: Pilot data and the GLUSENTIC study protocol.

A physiological feedback system exists between hepatocytes and the alpha cells, termed the liver-alpha cell axis and refers to the relationship between amino acid-stimulated glucagon secretion and glucagon-stimulated amino acid catabolism. Several reports indicate that non-alcoholic fatty liver disease (NAFLD) disrupts the liver-alpha cell axis, because of impaired glucagon receptor signaling (glucagon resistance). However, no experimental test exists to assess glucagon resistance in humans. The objective was to develop an experimental test to determine glucagon sensitivity with respect to amino acid and glucose metabolism in humans. The proposed glucagon sensitivity test (comprising two elements: 1) i.v. injection of 0.2 mg glucagon and 2) infusion of mixed amino acids 331 mg/hour/kg) is based on nine pilot studies which are presented. Calculation of a proposed glucagon sensitivity index with respect to amino acid catabolism is also described. Secondly, we describe a complete study protocol (GLUSENTIC) according to which the glucagon sensitivity test will be applied in a cross-sectional study currently taking place. 65 participants including 20 individuals with a BMI 18.6-25 kg/m2, 30 individuals with a BMI ≥ 25-40 kg/m2, and 15 individuals with type 1 diabetes with a BMI between 18.6 and 40 kg/m2 will be included. Participants will be grouped according to their degree of hepatic steatosis measured by whole-liver magnetic resonance imaging (MRI). The primary outcome measure will be differences in the glucagon sensitivity index between individuals with and without hepatic steatosis. Developing a glucagon sensitivity test and index may provide insight into the physiological and pathophysiological mechanism of glucagon action and glucagon-based therapies.

Chronic Sequelae After Muscle Strain Injuries: Influence of Heavy Resistance Training on Functional and Structural Characteristics in a Randomized Controlled Trial.

BACKGROUND: Muscle strain injury leads to a high risk of recurrent injury in sports and can cause long-term symptoms such as weakness and pain. Scar tissue formation after strain injuries has been described, yet what ultrastructural changes might occur in the chronic phase of this injury have not. It is also unknown if persistent symptoms and morphological abnormalities of the tissue can be mitigated by strength training. PURPOSE: To investigate if heavy resistance training improves symptoms and structural abnormalities after strain injuries. STUDY DESIGN: Randomized controlled trial; Level of evidence, 1. METHODS: A total of 30 participants with long-term weakness and/or pain after a strain injury of the thigh or calf muscles were randomized to eccentric heavy resistance training of the injured region or control exercises of the back and abdominal muscle. Isokinetic (hamstring) or isometric (calf) muscle strength was determined, muscle cross-sectional area measured, and pain and function evaluated. Scar tissue ultrastructure was determined from biopsy specimens taken from the injured area before and after the training intervention. RESULTS: Heavy resistance training over 3 months improved pain and function, normalized muscle strength deficits, and increased muscle cross-sectional area in the previously injured region. No systematic effect of training was found upon pathologic infiltration of fat and blood vessels into the previously injured area. Control exercises had no effect on strength, cross-sectional area, or scar tissue but a positive effect on patient-related outcome measures, such as pain and functional scores. CONCLUSION: Short-term strength training can improve sequelae symptoms and optimize muscle function even many years after a strain injury, but it does not seem to influence the overall structural abnormalities of the area with scar tissue. REGISTRATION: NCT02152098 (ClinicalTrials.gov identifier).

No Additive Clinical or Physiological Effects of Short-term Anti-inflammatory Treatment to Physical Rehabilitation in the Early Phase of Human Achilles Tendinopathy: A Randomized Controlled Trial.

BACKGROUND: Nonsteroidal anti-inflammatory drugs (NSAIDs) are commonly used in the treatment of Achilles tendinopathy, but whether they have any additive clinical effect on physical rehabilitation in the early phase of tendinopathy remains unknown. PURPOSE/HYPOTHESIS: To investigate whether an initial short-term NSAID treatment added to a physical rehabilitation program in the early phase of Achilles tendinopathy would have an additive effect. We hypothesized that the combination of NSAID and rehabilitation would be superior to rehabilitation alone. STUDY DESIGN: Randomized controlled trial; Level of evidence, 1. METHODS: A total of 69 patients with early phase Achilles tendinopathy (lasting <3 months) were randomly assigned to either a naproxen group (7 days of treatment; 500 mg twice daily; n = 34) or a placebo group (7 days of placebo treatment; n = 35). Both groups received an identical 12-week physical rehabilitation program. The clinical outcome of the study was evaluated using the Victorian Institute of Sports Assessment-Achilles (VISA-A) questionnaire and a numerical rating scale (NRS), and the physiological outcome was evaluated using ultrasonography, magnetic resonance imaging (MRI), and ultra-short time to echo T2* mapping MRI (UTE T2* MRI). Follow-up was performed at 1 week, 3 months, and 1 year. Time effects are presented as mean difference ± SEM. RESULTS: No significant differences were found between the 2 treatment groups for any of the outcome measures at any time point (P > .05). For the VISA-A score, a significant time effect was observed between baseline and 3-month follow-up (14.9 ± 2.3; P < .0001), and at 1-year follow-up, additional improvements were observed (6.1 ± 2.3; P < .01). Furthermore, the change in VISA-A score between baseline and 3-month follow-up was greater in patients with very short symptom duration (<1 month) at baseline compared with patients who had longer symptom duration (>2 months) (interaction between groups, 11.7 ± 4.2; P < .01). Despite clinical improvements, total weekly physical activity remained lower compared with preinjury levels at 3 months (-2.7 ± 0.5 h/wk; P < .0001) and 1 year (-3.0 ± 0.5 h/wk; P < .0001). At baseline, ultrasonography showed increased thickness (0.12 ± 0.03 cm; P < .0001) and vascularity (0.3 ± 0.1 cm2; P < .005) on the tendinopathic side compared with the contralateral side, but no changes over time were observed for ultrasonography, MRI, or UTE T2* MRI results. CONCLUSION: Clinical symptoms in early tendinopathy improved with physical rehabilitation, but this improvement was not augmented with the addition of NSAID treatment. Furthermore, this clinical recovery occurred in the absence of any measurable structural alterations. Finally, clinical improvements after a physical rehabilitation program were greater in patients with very short symptom duration compared with patients who had longer symptom duration. REGISTRATION: NCT03401177 (ClinicalTrials.gov identifier) and BFH-2016-019 (Danish Data Protection Agency).

The association between areas of secondary hyperalgesia and volumes of the caudate nuclei and other pain relevant brain structures-A 3-tesla MRI study of healthy men.

INTRODUCTION: Central sensitization plays a pivotal role in maintenance of pain and is believed to be intricately involved in several chronic pain conditions. One clinical manifestation of central sensitization is secondary hyperalgesia. The degree of secondary hyperalgesia presumably reflects individual levels of central sensitization. The objective of this study was to investigate the association between areas of secondary hyperalgesia and volumes of the caudate nuclei and other brain structures involved in pain processing. MATERIALS AND METHODS: We recruited 121 healthy male participants; 118 were included in the final analysis. All participants underwent whole brain magnetic resonance imaging (MRI). Prior to MRI, all participants underwent pain testing. Secondary hyperalgesia was induced by brief thermal sensitization. Additionally, we recorded heat pain detection thresholds (HPDT), pain during one minute thermal stimulation (p-TS) and results of the Pain Catastrophizing Scale (PCS) and Hospital Anxiety and Depression score (HADS). RESULTS: We found no significant associations between the size of the area of secondary hyperalgesia and the volume of the caudate nuclei or of the following structures: primary somatosensory cortex, anterior and mid cingulate cortex, putamen, nucleus accumbens, globus pallidus, insula and the cerebellum. Likewise, we found no significant associations between the volume of the caudate nuclei and HPDTs, p-TS, PCS and HADS. CONCLUSIONS: Our findings indicate that the size of the secondary hyperalgesia area is not associated with the volume of brain structures relevant for pain processing, suggesting that the propensity to develop central sensitization, assessed as secondary hyperalgesia, is not correlated to brain structure volume.