ABSTRACT
Background Radiology practices have a high volume of unremarkable chest radiographs and artificial intelligence (AI) could possibly improve workflow by providing an automatic report. Purpose To estimate the proportion of unremarkable chest radiographs, where AI can correctly exclude pathology (ie, specificity) without increasing diagnostic errors. Materials and Methods In this retrospective study, consecutive chest radiographs in unique adult patients (≥18 years of age) were obtained January 1-12, 2020, at four Danish hospitals. Exclusion criteria included insufficient radiology reports or AI output error. Two thoracic radiologists, who were blinded to AI output, labeled chest radiographs as "remarkable" or "unremarkable" based on predefined unremarkable findings (reference standard). Radiology reports were classified similarly. A commercial AI tool was adapted to output a chest radiograph "remarkableness" probability, which was used to calculate specificity at different AI sensitivities. Chest radiographs with missed findings by AI and/or the radiology report were graded by one thoracic radiologist as critical, clinically significant, or clinically insignificant. Paired proportions were compared using the McNemar test. Results A total of 1961 patients were included (median age, 72 years [IQR, 58-81 years]; 993 female), with one chest radiograph per patient. The reference standard labeled 1231 of 1961 chest radiographs (62.8%) as remarkable and 730 of 1961 (37.2%) as unremarkable. At 99.9%, 99.0%, and 98.0% sensitivity, the AI had a specificity of 24.5% (179 of 730 radiographs [95% CI: 21, 28]), 47.1% (344 of 730 radiographs [95% CI: 43, 51]), and 52.7% (385 of 730 radiographs [95% CI: 49, 56]), respectively. With the AI fixed to have a similar sensitivity as radiology reports (87.2%), the missed findings of AI and reports had 2.2% (27 of 1231 radiographs) and 1.1% (14 of 1231 radiographs) classified as critical (P = .01), 4.1% (51 of 1231 radiographs) and 3.6% (44 of 1231 radiographs) classified as clinically significant (P = .46), and 6.5% (80 of 1231) and 8.1% (100 of 1231) classified as clinically insignificant (P = .11), respectively. At sensitivities greater than or equal to 95.4%, the AI tool exhibited less than or equal to 1.1% critical misses. Conclusion A commercial AI tool used off-label could correctly exclude pathology in 24.5%-52.7% of all unremarkable chest radiographs at greater than or equal to 98% sensitivity. The AI had equal or lower rates of critical misses than radiology reports at sensitivities greater than or equal to 95.4%. These results should be confirmed in a prospective study. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Yoon and Hwang in this issue.
Subject(s)
Artificial Intelligence , Radiography, Thoracic , Humans , Radiography, Thoracic/methods , Female , Aged , Male , Retrospective Studies , Middle Aged , Aged, 80 and over , Sensitivity and Specificity , Denmark , Diagnostic Errors/statistics & numerical dataABSTRACT
Background Due to conflicting findings in the literature, there are concerns about a lack of objectivity in grading knee osteoarthritis (KOA) on radiographs. Purpose To examine how artificial intelligence (AI) assistance affects the performance and interobserver agreement of radiologists and orthopedists of various experience levels when evaluating KOA on radiographs according to the established Kellgren-Lawrence (KL) grading system. Materials and Methods In this retrospective observer performance study, consecutive standing knee radiographs from patients with suspected KOA were collected from three participating European centers between April 2019 and May 2022. Each center recruited four readers across radiology and orthopedic surgery at in-training and board-certified experience levels. KL grading (KL-0 = no KOA, KL-4 = severe KOA) on the frontal view was assessed by readers with and without assistance from a commercial AI tool. The majority vote of three musculoskeletal radiology consultants established the reference standard. The ordinal receiver operating characteristic method was used to estimate grading performance. Light kappa was used to estimate interrater agreement, and bootstrapped t statistics were used to compare groups. Results Seventy-five studies were included from each center, totaling 225 studies (mean patient age, 55 years ± 15 [SD]; 113 female patients). The KL grades were KL-0, 24.0% (n = 54); KL-1, 28.0% (n = 63); KL-2, 21.8% (n = 49); KL-3, 18.7% (n = 42); and KL-4, 7.6% (n = 17). Eleven readers completed their readings. Three of the six junior readers showed higher KL grading performance with versus without AI assistance (area under the receiver operating characteristic curve, 0.81 ± 0.017 [SEM] vs 0.88 ± 0.011 [P < .001]; 0.76 ± 0.018 vs 0.86 ± 0.013 [P < .001]; and 0.89 ± 0.011 vs 0.91 ± 0.009 [P = .008]). Interobserver agreement for KL grading among all readers was higher with versus without AI assistance (κ = 0.77 ± 0.018 [SEM] vs 0.85 ± 0.013; P < .001). Board-certified radiologists achieved almost perfect agreement for KL grading when assisted by AI (κ = 0.90 ± 0.01), which was higher than that achieved by the reference readers independently (κ = 0.84 ± 0.017; P = .01). Conclusion AI assistance increased junior readers' radiographic KOA grading performance and increased interobserver agreement for osteoarthritis grading across all readers and experience levels. Published under a CC BY 4.0 license. Supplemental material is available for this article.
Subject(s)
Artificial Intelligence , Observer Variation , Osteoarthritis, Knee , Humans , Female , Male , Osteoarthritis, Knee/diagnostic imaging , Middle Aged , Retrospective Studies , Radiography/methods , AgedABSTRACT
PURPOSE: To estimate the ability of a commercially available artificial intelligence (AI) tool to detect acute brain ischemia on Magnetic Resonance Imaging (MRI), compared to an experienced neuroradiologist. METHODS: We retrospectively included 1030 patients with brain MRI, suspected of stroke from January 6th, 2020 to 1st of April 2022, based on these criteria: Age ≥ 18 years, symptoms within four weeks before the scan. The neuroradiologist reinterpreted the MRI scans and subclassified ischemic lesions for reference. We excluded scans with interpretation difficulties due to artifacts or missing sequences. Four MRI scanner models from the same vendor were used. The first 800 patients were included consecutively, remaining enriched for less frequent lesions. The index test was a CE-approved AI tool (Apollo version 2.1.1 by Cerebriu). RESULTS: The final analysis cohort comprised 995 patients (mean age 69 years, 53 % female). A case-based analysis for detecting acute ischemic lesions showed a sensitivity of 89 % (95 % CI: 85 %-91 %) and specificity of 90 % (95 % CI: 87 %-92 %). We found no significant difference in sensitivity or specificity based on sex, age, or comorbidities. Specificity was reduced in cases with DWI artifacts. Multivariate analysis showed that increasing ischemic lesion size and fragmented lesions were independently associated with higher sensitivity, while non-acute lesion ages lowered sensitivity. CONCLUSIONS: The AI tool exhibits high sensitivity and specificity in detecting acute ischemic lesions on MRI compared to an experienced neuroradiologist. While sensitivity depends on the ischemic lesions' characteristics, specificity depends on the image quality.
Subject(s)
Brain Ischemia , Deep Learning , Stroke , Humans , Female , Aged , Adolescent , Male , Retrospective Studies , Artificial Intelligence , Stroke/pathology , Magnetic Resonance Imaging/methods , Brain Ischemia/diagnostic imaging , Brain Ischemia/pathology , Brain/pathology , Algorithms , Diagnostic Tests, Routine , Diffusion Magnetic Resonance Imaging/methodsABSTRACT
Background Commercially available artificial intelligence (AI) tools can assist radiologists in interpreting chest radiographs, but their real-life diagnostic accuracy remains unclear. Purpose To evaluate the diagnostic accuracy of four commercially available AI tools for detection of airspace disease, pneumothorax, and pleural effusion on chest radiographs. Materials and Methods This retrospective study included consecutive adult patients who underwent chest radiography at one of four Danish hospitals in January 2020. Two thoracic radiologists (or three, in cases of disagreement) who had access to all previous and future imaging labeled chest radiographs independently for the reference standard. Area under the receiver operating characteristic curve, sensitivity, and specificity were calculated. Sensitivity and specificity were additionally stratified according to the severity of findings, number of findings on chest radiographs, and radiographic projection. The χ2 and McNemar tests were used for comparisons. Results The data set comprised 2040 patients (median age, 72 years [IQR, 58-81 years]; 1033 female), of whom 669 (32.8%) had target findings. The AI tools demonstrated areas under the receiver operating characteristic curve ranging 0.83-0.88 for airspace disease, 0.89-0.97 for pneumothorax, and 0.94-0.97 for pleural effusion. Sensitivities ranged 72%-91% for airspace disease, 63%-90% for pneumothorax, and 62%-95% for pleural effusion. Negative predictive values ranged 92%-100% for all target findings. In airspace disease, pneumothorax, and pleural effusion, specificity was high for chest radiographs with normal or single findings (range, 85%-96%, 99%-100%, and 95%-100%, respectively) and markedly lower for chest radiographs with four or more findings (range, 27%-69%, 96%-99%, 65%-92%, respectively) (P < .001). AI sensitivity was lower for vague airspace disease (range, 33%-61%) and small pneumothorax or pleural effusion (range, 9%-94%) compared with larger findings (range, 81%-100%; P value range, > .99 to < .001). Conclusion Current-generation AI tools showed moderate to high sensitivity for detecting airspace disease, pneumothorax, and pleural effusion on chest radiographs. However, they produced more false-positive findings than radiology reports, and their performance decreased for smaller-sized target findings and when multiple findings were present. © RSNA, 2023 Supplemental material is available for this article. See also the editorial by Yanagawa and Tomiyama in this issue.
Subject(s)
Deep Learning , Pleural Effusion , Pneumothorax , Adult , Humans , Female , Aged , Artificial Intelligence , Pneumothorax/diagnostic imaging , Retrospective Studies , Radiography, Thoracic/methods , Sensitivity and Specificity , Pleural Effusion/diagnostic imagingABSTRACT
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.
Subject(s)
Artificial Intelligence , Radiography, Thoracic , Adult , Humans , Female , Aged , Retrospective Studies , Radiography, Thoracic/methods , Radiographic Image Interpretation, Computer-Assisted/methods , RadiologistsABSTRACT
BACKGROUND: Colchicine has been suggested for osteoarthritis treatment, but evidence is contradictory. We aimed to investigate colchicine's efficacy and safety compared with placebo in people with hand osteoarthritis. METHODS: In this single-centre, double-blind, randomised, placebo-controlled trial we recruited adults with symptomatic hand osteoarthritis and finger pain of at least 40 mm on a 100 mm visual analogue scale from an outpatient clinic in Denmark. The hand with the most severe finger pain at inclusion was the target hand. Participants were randomly assigned (1:1) to 0·5 mg colchicine or placebo taken orally twice a day for 12 weeks, stratified by BMI (≥30 kg/m2), sex, and age (≥75 years). Participants, outcome assessors, and data analysts were masked to treatment allocation. The primary endpoint was change from baseline to week 12 in target hand finger pain, assessed on a 100 mm visual analogue scale with a pre-specified minimal clinically important difference of 15 mm, in the intention-to-treat population. Safety was assessed at week 12 in the intention-to-treat population. The study was registered with ClinicalTrials.gov, NCT04601883, and with EudraCT, 2020-002803-20. FINDINGS: Between Jan 15, 2021, and March 3, 2022, 186 people were screened for eligibility, and 100 were randomly assigned to receive colchicine (n=50) or placebo (n=50). Participants had a mean age of 70·9 (SD 7·5) years, 69 (69%) of 100 were women and 31 (31%) were men. All participants completed the study. The mean change from baseline to week 12 in finger pain were -13·9 mm (SE 2·8) in the colchicine group and -13·5 mm (2·8) in the placebo group, with a between-group difference (colchicine vs placebo) of -0·4 mm (95% CI -7·6 to 6·7; p=0·90). In the colchicine group, there were 76 adverse events in 36 (72%) of 50 participants and one serious adverse advent (migraine attack leading to hospital admission). In the placebo group, there were 42 adverse events in 22 (44%) of 50 participants and two serious adverse events (cholecystitis and elevated alanine aminotransferase concentrations, in the same patient). INTERPRETATION: In people with painful hand osteoarthritis, treatment with 0·5 mg of colchicine twice day for 12 weeks did not effectively relieve pain, and treatment with colchicine was associated with more adverse events. FUNDING: The Oak Foundation, IMK Almene Fond, Minister Erna Hamilton's Scholarship for Science and Art, AP Moller and Wife Chastine McKinney Moller's Foundation for Medical Science Advancement, The Danish Medical Association, the Velux Foundation, Aase and Ejnar Danielsen's Foundation, and Director Emil C Hertz and Wife Inger Hertz's foundation.
Subject(s)
Hand , Upper Extremity , Adult , Male , Humans , Female , Aged , Double-Blind Method , Colchicine/adverse effects , PainABSTRACT
BACKGROUND: The main challenge with fat grafting is loss of some of the graft to postsurgery resorption. Previous studies suggest that adipose-derived stromal cells (ASCs) can improve the volume retention of fat grafts but there is a lack of randomized trials to support the use of ASCs in clinical practice. OBJECTIVES: This trial aimed to investigate whether ASCs improve fat graft volume retention in patients undergoing breast augmentation with lipofilling. METHODS: This was a double-blind, randomized controlled trial of breast augmentation with ASC-enriched fat grafting. Healthy women aged 30 to 45 years were enrolled. First, the participants underwent liposuction to obtain fat for culture expansion of ASCs. Then, the participants were randomly assigned to undergo a 300- to 350-mL breast augmentation with ASC-enriched fat grafting (10â ×â 106 ASCs/mL fat graft) to 1 of their breasts and placebo-enriched fat grafting of identical volume to the contralateral breast. The primary outcome was fat graft volume retention after a 1-year follow-up measured with MRI. The trial is registered at www.clinicaltrialsregister.eu (EudraCT-2014-000510-59). RESULTS: Ten participants were included in the trial; all completed the treatment and follow-up. No serious adverse events occurred. Fat graft volume retention after 1 year was 54.0% (95% CI, 30.4%-77.6%) in the breasts treated with ASC-enriched fat grafting (nâ =â 10) and 55.9% (95% CI, 28.9%-82.9%) in the contralateral breasts treated with placebo-enriched fat grafting (nâ =â 10) (Pâ =â 0.566). CONCLUSIONS: The findings of this trial do not support that ASC-enriched fat grafting is superior to standard fat grafting for breast augmentation.
Subject(s)
Lipectomy , Mammaplasty , Mesenchymal Stem Cell Transplantation , Adipose Tissue/transplantation , Female , Humans , Stromal Cells/transplantationABSTRACT
PURPOSE: The primary aim was to investigate the diagnostic performance of an Artificial Intelligence (AI) algorithm for pneumoperitoneum detection in patients with acute abdominal pain who underwent an abdominal CT scan. METHOD: This retrospective diagnostic test accuracy study used a consecutive patient cohort from the Acute High-risk Abdominal patient population at Herlev and Gentofte Hospital, Denmark between January 1, 2019 and September 25, 2019. As reference standard, all studies were rated for pneumoperitoneum (subgroups: none, small, medium, and large amounts) by a gastrointestinal radiology consultant. The index test was a novel AI algorithm based on a sliding window approach with a deep recurrent neural network at its core. The primary outcome was the area under the curve (AUC) of the receiver operating characteristic (ROC). RESULTS: Of 331 included patients (median age 68 years (Range 19-100; 180 women)) 31 patients (9%) had pneumoperitoneum (large: 16, moderate: 7, small: 8). The AUC was 0.77 (95% CI 0.66-0.87). At a specificity of 99% (297/300, 95% CI: 97-100%), sensitivity was 52% (16/31, 95% CI 29-65%), and positive likelihood ratio was 52 (95% CI 16-165). When excluding cases with smaller amounts of free air (<0.25 mL) the AUC increased to 0.96 (95% CI 0.89-1.0). At 99% specificity, sensitivity was 81% (13/16) and positive likelihood ratio was 82 (95% CI 27 - 254). CONCLUSIONS: An AI algorithm identified pneumoperitoneum on CT scans in a clinical setting with low sensitivity but very high specificity, supporting its role for ruling in pneumoperitoneum.
Subject(s)
Abdomen, Acute , Pneumoperitoneum , Abdominal Pain/diagnostic imaging , Abdominal Pain/etiology , Adult , Aged , Aged, 80 and over , Artificial Intelligence , Diagnostic Tests, Routine , Female , Humans , Middle Aged , Pneumoperitoneum/diagnostic imaging , Retrospective Studies , Tomography, X-Ray Computed , Young AdultABSTRACT
RATIONAL AND OBJECTIVES: This study investigated how an AI tool impacted radiologists reading time for non-contrast chest CT exams. MATERIALS AND METHODS: An AI tool was implemented into the PACS reading workflow of non-contrast chest CT exams between April and May 2020. The reading time was recorded for one CONSULTANT RADIOLOGIST and one RADIOLOGY RESIDENT by an external observer. After each case radiologists answered questions regarding additional findings and perceived case overview. Reading times were recorded for 25 cases without and 20 cases with AI tool assistance for each reader. Differences in reading time with and without the AI tool were assessed using Welch's t-test for non-inferiority with non-inferiority limits defined as 100 seconds for the consultant and 200 seconds for the resident. RESULTS: The mean reading time for the radiology resident was not significantly affected by the AI tool (without AI 370s vs with AI 437s; +67s 95% CI -28s to +163s, p = 0.16). The reading time for the radiology consultant was also not significantly affected by the AI tool (without AI 366s vs with AI 380s; +13s (95% CI - -57s to 84s, p = 0.70). The AI tool led to additional actionable findings in 5/40 (12.5%) studies and better overview in 18/20 (90%) of studies for the resident. CONCLUSION: A PACS based implementation of an AI tool for concurrent reading of chest CT exams did not increase reading time with additional actionable findings made as well as a perceived better case overview for the radiology resident.
Subject(s)
Artificial Intelligence , Radiology , Feasibility Studies , Humans , Prospective Studies , RadiologistsABSTRACT
PURPOSE: Purpose of this study was to investigate Dual-energy CT (DECT) derived virtual non-calcium (VNCa) values for absolute quantification of the bone marrow composition in the wrist. MATERIALS AND METHODS: We prospectively included consecutive adult participants and examined their wrists with DECT. Ranges of VNCa and calcium values were measured in the carpal bones, radius and ulna using a semi-automatic method. Bones with bone marrow edema, assessed by two blinded radiologists, were excluded. After determining optimum parameters for the three-material decomposition, the influence of calcium values, age and sex on VNCa values was assessed using multiple linear regression. RESULTS: 41 participants (Median age 53 years, range 20 years - 88 years, 51 % men) were enrolled and 399 bones assessed. At participant level mean VNCa values were -143 HU (SD 14 HU) using the current parameters for three-material decomposition and -104HU (SD 11 HU) with optimized parameters. There was a strong and significant influence of calcium values on VNCa values with the current parameters (pâ¯<â¯0.001, -0.137 HU[VNCa] / HU[Calcium]). With optimized parameters the calcium values and sex were not statistically significant predictors of VNCa values. Age was a significant, but clinically negligible, predictor (pâ¯=â¯0.03, -0.225 HU / year). CONCLUSIONS: After optimizing three-material decomposition parameters, calcium values, age and sex do not substantially influence virtual non-calcium values, and DECT may therefore be used for absolute quantification of the bone marrow composition - alleviating the need for reference bones or groups.
Subject(s)
Bone Marrow , Wrist , Adult , Bone Marrow/diagnostic imaging , Female , Humans , Magnetic Resonance Imaging , Male , Sensitivity and Specificity , Tomography, X-Ray Computed , Young AdultABSTRACT
Background Patients with wrist trauma and negative findings on radiographs often undergo additional MRI examinations to assess for radiographically occult fractures. Dual-energy CT may be more readily available than MRI in some settings. Purpose To evaluate the diagnostic test accuracy of dual-energy CT in helping detect bone marrow edema and fracture in participants with wrist trauma and clinical suspicion of a wrist fracture but with negative findings on radiographs. Materials and Methods Adults were prospectively enrolled between January 2018 and November 2018. Wrists were examined with dual-energy CT and MRI, and images were read by four readers who were blinded to clinical information. The presence of bone marrow edema and fracture was rated per bone. The reference standard for bone marrow edema was the combined reading of MRI scans. The reference standard for fracture was a combined reading of MRI and dual-energy CT scans. A fifth radiologist arbitrated results in case of discrepancies. Diagnostic test accuracy was calculated per reader and for readers combined using exact binomial tests. Results Forty-six participants (mean age, 47 years ± 19 [standard deviation]; 24 men [52%]) were enrolled, and 750 bones (50 wrists) were assessed. Dual-energy CT had an average sensitivity of 94% (95% confidence interval [CI]: 80%, 99%; 31 of 33 wrists) and specificity of 65% (95% CI: 38%, 86%; 11 of 17 wrists) in the detection of wrists with bone marrow edema and a sensitivity of 69% (95% CI: 55%, 81%; 36 of 52 bones) and a specificity of 98% (95% CI: 97%, 99%; 682 of 696 bones) in the detection of edema in individual bones. MRI had a sensitivity of 80% (95% CI: 63%, 91%; 28 of 35 wrists) and a specificity of 93% (95% CI: 68%, 100%; 14 of 15 wrists) in helping detect wrists with fractures. Dual-energy CT had a sensitivity of 91% (95% CI: 77%, 98%; 32 of 35 wrists) and a specificity of 87% (95% CI: 60%, 98%; 53 of 60 wrists) in helping detect wrists with fractures. McNemar tests showed no significant differences between MRI and dual-energy CT (P = .07 to >.99) for all readers. Conclusion Dual-energy CT had a high sensitivity and a moderate specificity in the detection of bone marrow edema of the wrist. Dual-energy CT had high sensitivity and specificity in depicting fractures of the wrist in patients with suspected wrist fractures and negative findings on radiographs. © RSNA, 2020 Online supplemental material is available for this article. See also the editorial by Fukuda in this issue.
Subject(s)
Fractures, Bone/diagnostic imaging , Radiography, Dual-Energy Scanned Projection/methods , Tomography, X-Ray Computed/methods , Wrist Injuries/diagnostic imaging , Adult , Female , Humans , Male , Middle Aged , Prospective Studies , Sensitivity and Specificity , Wrist/diagnostic imagingABSTRACT
BACKGROUND: We investigated the influence of dose, spectral separation, pitch, rotation time, and reconstruction kernel on accuracy and image noise of virtual non-calcium images using a bone marrow phantom. METHODS: The phantom was developed at our institution and scanned using a third-generation dual-source dual-energy CT scanner at five different spectral separations by varying the tube-voltage combinations (70 kV/Sn150 kV, 80 kV/Sn150 kV, 90 kV/Sn150 kV, and 100 kV/Sn150 kV, all with 0.6-mm tin filter [Sn]; 80 kV/140 kV without tin filter) at six different doses (volume computed tomography dose index from 1 to 80 mGy). In separate experiments, rotation times, pitch, and reconstruction kernels were varied at a constant dose and tube voltage. Accuracy was determined by measuring the mean error between virtual non-calcium values in the fluid within and outside of the bone. Image noise was defined as the standard deviation of virtual non-calcium values. RESULTS: Spectral separation, dose, rotation time, or pitch did not significantly correlate (p > 0.083) with mean error. Increased spectral separation (rs-0.96, p < 0.001) and increased dose (rs-0.98, p < 0.001) correlated significantly with decreased image noise. Increasing sharpness of the reconstruction kernel correlated with mean error (rs 0.83, p = 0.015) and image noise (rs 1.0, p < 0.001). CONCLUSIONS: Increased dose and increased spectral separation significantly lowered image noise in virtual non-calcium images but did not affect the accuracy. Virtual non-calcium reconstructions with similar accuracy and image noise could be achieved at a lower tube-voltage difference by increasing the dose.
Subject(s)
Bone Marrow/diagnostic imaging , Tomography, X-Ray Computed/methods , Phantoms, Imaging , Prospective Studies , Tomography, X-Ray Computed/instrumentationSubject(s)
Cardiopulmonary Resuscitation , Heart Arrest , Emergency Service, Hospital , Humans , Machine LearningABSTRACT
Several techniques for measuring breast volume (BV) are based on examining the breast on magnetic resonance imaging. However, when techniques designed to measure total BV are used to quantify BV changes, for example, after fat grafting, a systematic error is introduced because BV changes lead to contour alterations of the breast. The volume of the altered breast includes not only the injected volume but also tissue previously surrounding the breast. Therefore, the quantitative difference in BV before and after augmentation will differ from the injected volume. Here, we present a new technique to measure BV changes that compensates for this systematic error by defining the boundaries of the breast to immovable osseous pointers. This approach avoids the misinterpretation of tissue included within the expanded boundaries as graft tissue. This new method of analysis may be a reliable tool for assessing BV changes to determine fat graft retention and may be useful for evaluating and comparing available surgical techniques for breast augmentation and reconstruction using fat grafting.
ABSTRACT
BACKGROUND: Biocytin has found numerous uses as a neuronal tracer, since it shows both antero- and retrograde transport in neuronal tracts. The main advantage of biocytin lies in the comprehensive intracellular distribution of the molecule, and in effective detection using avidin-based reactions. The main drawback is that biocytin cannot be visualized in live tissue. NEW METHOD: We demonstrate that TMR biocytin, a conjugate of biocytin and a rhodamine fluorophore, is an effective neuronal tracer in live tissue when applied by electroporation. RESULTS: The initial fiber transport velocity of TMR biocytin is high-5.4mm/h. TMR biocytin can be used in conjunction with AM calcium dyes to label neuronal somas from distances of several millimetres, and record calcium transients during the course of a few hours. Juxtacellular application of TMR biocytin leads to fast anterograde transport with labeling of local synapses within 10min. TMR biocytin is fixable, stable during methyl salicylate clearing, and can be visualized deep in nervous tissue. COMPARISON WITH EXISTING METHODS: Retrograde labeling with TMR biocytin enables long-range neuronal visualization and concurrent calcium imaging after only a few hours, which is substantially faster than other fluorescence-based tracers. The green emitting Atto 488 biotin is also taken up and transported retrogradely, but it is not compatible with standard green emitting calcium dyes. CONCLUSIONS: TMR biocytin is an attractive neuronal tracer. It labels neurons fast over long distances, and it can be used in conjunction with calcium dyes to report on neuronal activity in retrogradely labeled live neurons.
Subject(s)
Brain Stem/cytology , Fluorescent Dyes/metabolism , Lysine/analogs & derivatives , Neural Pathways/physiology , Neurons/metabolism , Analysis of Variance , Aniline Compounds/metabolism , Animals , Animals, Newborn , Calcium , Electroporation , In Vitro Techniques , Lysine/metabolism , Mice , Rhodamines/metabolism , Xanthenes/metabolismABSTRACT
The inferior olivary nucleus (IO) in in vitro slices from postnatal mice (P5.5-P15.5) spontaneously generates clusters of neurons with synchronous calcium transients, and intracellular recordings from IO neurons suggest that electrical coupling between neighbouring IO neurons may serve as a synchronizing mechanism. Here, we studied the cluster-forming mechanism and find that clusters overlap extensively with an overlap distribution that resembles the distribution for a random overlap model. The average somatodendritic field size of single curly IO neurons was â¼6400 µm(2), which is slightly smaller than the average IO cluster size. Eighty-seven neurons with overlapping dendrites were estimated to be contained in the principal olive mean cluster size, and about six non-overlapping curly IO neurons could be contained within the largest clusters. Clusters could also be induced by iontophoresis with glutamate. Induced clusters were inhibited by tetrodotoxin, carbenoxelone and 18ß-glycyrrhetinic acid, suggesting that sodium action potentials and electrical coupling are involved in glutamate-induced cluster formation, which could also be induced by activation of N-methyl-d-aspartate and α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptors. Spikelets and a small transient depolarizing response were observed during glutamate-induced cluster formation. Calcium transients spread with decreasing velocity during cluster formation, and somatic action potentials and cluster formation are accompanied by large dendritic calcium transients. In conclusion, cluster formation depends on gap junctions, sodium action potentials and spontaneous clusters occur randomly throughout the IO. The relative slow signal spread during cluster formation, combined with a strong dendritic influx of calcium, may signify that active dendritic properties contribute to cluster formation.