Radiological Interpretation Research Articles

Construction and Application of Rib Fracture Diagnosis Model Based on YOLOv3 Algorithm.

The artificial intelligence-aided diagnosis model of rib fractures based on YOLOv3 algorithm was established and applied to practical case to explore the application advantages in rib fracture cases in forensic medicine. DICOM format CT images of 884 cases with rib fractures caused by thoracic trauma were collected, and 801 of them were used as training and validation sets. A rib fracture diagnosis model based on YOLOv3 algorithm and Darknet53 as the backbone network was built. After the model was established, 83 cases were taken as the test set, and the precision rate, recall rate, F1-score and radiology interpretation time were calculated. The model was used to diagnose a practical case and compared with manual diagnosis. The established model was used to test 83 cases, the fracture precision rate of this model was 90.5%, the recall rate was 75.4%, F1-score was 0.82, the radiology interpretation time was 4.4 images per second and the identification time of each patient's data was 21 s, much faster than manual diagnosis. The recognition results of the model was consistent with that of the manual diagnosis. The rib fracture diagnosis model in practical case based on YOLOv3 algorithm can quickly and accurately identify fractures, and the model is easy to operate. It can be used as an auxiliary diagnostic technique in forensic clinical identification.

The Role of Radiology in Multidisciplinary Discussion of Patients With Interstitial Lung Diseases.

Radiologists fulfill a vital role in the multidisciplinary care provided to patients with interstitial lung diseases and other diffuse parenchymal lung disorders. The diagnosis of interstitial lung diseases hinges on the consensus of clinical, radiology, and pathology medical subspecialists, but additional expertise from rheumatology, immunology, or hematology can be invaluable. The thin-section computed tomography (CT) features of lung involvement informs the diagnostic approach. Radiologists should be familiar with radiologic methods (including inspiratory/expiratory and prone imaging) and be well versed in the recognition of the CT features of fibrosis, assessment of the overall pattern of lung involvement, and classification according to the latest guidelines. We present a case-based review that highlights examples wherein CT features and subspecialist radiologist interpretation informed the multidisciplinary team consensus diagnosis and care pathways.

Effects of a comprehensive brain computed tomography deep learning model on radiologist detection accuracy

ObjectivesNon-contrast computed tomography of the brain (NCCTB) is commonly used to detect intracranial pathology but is subject to interpretation errors. Machine learning can augment clinical decision-making and improve NCCTB scan interpretation. This retrospective detection accuracy study assessed the performance of radiologists assisted by a deep learning model and compared the standalone performance of the model with that of unassisted radiologists.MethodsA deep learning model was trained on 212,484 NCCTB scans drawn from a private radiology group in Australia. Scans from inpatient, outpatient, and emergency settings were included. Scan inclusion criteria were age ≥ 18 years and series slice thickness ≤ 1.5 mm. Thirty-two radiologists reviewed 2848 scans with and without the assistance of the deep learning system and rated their confidence in the presence of each finding using a 7-point scale. Differences in AUC and Matthews correlation coefficient (MCC) were calculated using a ground-truth gold standard.ResultsThe model demonstrated an average area under the receiver operating characteristic curve (AUC) of 0.93 across 144 NCCTB findings and significantly improved radiologist interpretation performance. Assisted and unassisted radiologists demonstrated an average AUC of 0.79 and 0.73 across 22 grouped parent findings and 0.72 and 0.68 across 189 child findings, respectively. When assisted by the model, radiologist AUC was significantly improved for 91 findings (158 findings were non-inferior), and reading time was significantly reduced.ConclusionsThe assistance of a comprehensive deep learning model significantly improved radiologist detection accuracy across a wide range of clinical findings and demonstrated the potential to improve NCCTB interpretation.Clinical relevance statementThis study evaluated a comprehensive CT brain deep learning model, which performed strongly, improved the performance of radiologists, and reduced interpretation time. The model may reduce errors, improve efficiency, facilitate triage, and better enable the delivery of timely patient care.Key Points• This study demonstrated that the use of a comprehensive deep learning system assisted radiologists in the detection of a wide range of abnormalities on non-contrast brain computed tomography scans.• The deep learning model demonstrated an average area under the receiver operating characteristic curve of 0.93 across 144 findings and significantly improved radiologist interpretation performance.• The assistance of the comprehensive deep learning model significantly reduced the time required for radiologists to interpret computed tomography scans of the brain.

Ultrasound-guided injection through the rotator cuff interval: a clinical perspective of one institution's results and description of technique.

Fluoroscopic injection through the rotator cuff interval (RCI) is a common technique for diagnostic arthrography and therapeutic intervention. Ultrasound approaches through the RCI have been less commonly studied, but there is a growing body of literature. The purpose of this study was to present a standardized technique of ultrasound-guided injection into the glenohumeral joint utilizing the RCI in magnetic resonance imaging (MRI) arthrography (MRA) and to report one medical group's experience with the technique. A retrospective chart review of all ultrasound-guided injections into the glenohumeral joint utilizing the RCI was performed from July 1, 2014 through June 1, 2021. Data were compiled for age, gender, body mass index (BMI), and prior surgery on the shoulder. The primary endpoint was successful administration of intra-articular dilute gadolinium contrast adequate for radiologic interpretation. A total of 487 injections into the glenohumeral joint via the RCI were performed. One hundred and fifty-five patients had previous shoulder surgery, with the remainder naive to intervention. The success rate of injections into the glenohumeral joint was 99.4 %, with only three injections considered unsuccessful. The three unsuccessful injections did not succeed because of a lack of intra-articular contrast media present. This success rate is impressive and promising, particularly when considering that 155 of the patients had previous surgery, which could potentially cause complications, and because these injections were performed over a long period of 7 years. Accessing the RCI under ultrasound guidance is a very successful technique for injection within the glenohumeral joint.

Teleradiology Service is Indispensable in the Indian Healthcare Sector

Teleradiology is an indispensable service in the healthcare sector, offering numerous benefits to healthcare providers, patients, and radiologists. An ideal teleradiology service provider is one with a sturdy infrastructure and technology provisions with a plethora of highly qualified board-certified radiologists who would cater to the needs of accurate, secure, quality, and timely radiologic interpretations with 24/7/365 coverage for small hospitals, clinics, diagnostic imaging centers and emergency rooms. The aim of the article is to give perspective on teleradiology service as an indispensable service in the Indian healthcare sector. The article also emphasizes the key elements an ideal teleradiology service provider should have.

Can deep learning replace histopathological examinations in the differential diagnosis of cervical lymphadenopathy?

We aimed to develop a diagnostic deep learning model using contrast-enhanced CT images and to investigate whether cervical lymphadenopathies can be diagnosed with these deep learning methods without radiologist interpretations and histopathological examinations. A total of 400 patients who underwent surgery for lymphadenopathy in the neck between 2010 and 2022 were retrospectively analyzed. They were examined in four groups of 100 patients: the granulomatous diseases group, the lymphoma group, the squamous cell tumor group, and the reactive hyperplasia group. The diagnoses of the patients were confirmed histopathologically. Two CT images from all the patients in each group were used in the study. The CT images were classified using ResNet50, NASNetMobile, and DenseNet121 architecture input. The classification accuracies obtained with ResNet50, DenseNet121, and NASNetMobile were 92.5%, 90.62, and 87.5, respectively. Deep learning is a useful diagnostic tool in diagnosing cervical lymphadenopathy. In the near future, many diseases could be diagnosed with deep learning models without radiologist interpretations and invasive examinations such as histopathological examinations. However, further studies with much larger case series are needed to develop accurate deep-learning models.

Absence of Horizontal Fissure, Middle Lobe Fusion and Anomalous Appearance of Mediastinal Surface of Right Lungs: A Rare Case Report

Anatomical variations of lungs in the form of presence of any extra lobe or absence of anatomical lobe, fissures of lungs are important to know to avoid injuries during surgeries related to lungs and associated structures. In this case report absence of middle lobe and horizontal fissure in two right lungs extracted from male cadaver during routine dissection in the Anatomy department, Adesh institute of medical sciences and research, Bathinda, Punjab, India was noted and reported. Every anatomical structure in left side lung of same cadaver was found normal. Also in one abnormal right lung, gross appearance of right lung was looked like left lung. Anatomical knowledge and statistics of every type of variations are important for anatomists for teaching purposes and for surgeons while performing lobectomies, surgical resections of segments of lungs and for radiological interpretations.

Analysis of Specimen Mammography with Artificial Intelligence to Predict Margin Status.

Intraoperative specimen mammography is a valuable tool in breast cancer surgery, providing immediate assessment of margins for a resected tumor. However, the accuracy of specimen mammography in detecting microscopic margin positivity is low. We sought to develop an artificial intelligence model to predict the pathologic margin status of resected breast tumors using specimen mammography. A dataset of specimen mammography images matched with pathologic margin status was collected from our institution from 2017 to 2020. The dataset was randomly split into training, validation, and test sets. Specimen mammography models pretrained on radiologic images were developed and compared with models pretrained on nonmedical images. Model performance was assessed using sensitivity, specificity, and area under the receiver operating characteristic curve (AUROC). The dataset included 821 images, and 53% had positive margins. For three out of four model architectures tested, models pretrained on radiologic images outperformed nonmedical models. The highest performing model, InceptionV3, showed sensitivity of 84%, specificity of 42%, and AUROC of 0.71. Model performance was better among patients with invasive cancers, less dense breasts, and non-white race. This study developed and internally validated artificial intelligence models that predict pathologic margins status for partial mastectomy from specimen mammograms. The models' accuracy compares favorably with published literature on surgeon and radiologist interpretation of specimen mammography. With further development, these models could more precisely guide the extent of resection, potentially improving cosmesis and reducing reoperations.

Comparing Predictive Utility of Head Computed Tomography Scan-Based Scoring Systems for Traumatic Brain Injury: A Retrospective Study.

This study compared the predictive utility of Marshall, Rotterdam, Stockholm, Helsinki, and NeuroImaging Radiological Interpretation System (NIRIS) scorings based on early non-contrast brain computed tomography (CT) scans in patients with traumatic brain injury (TBI). The area under a receiver operating characteristic curve (AUROC) was used to determine the predictive utility of scoring systems. Subgroup analyses were performed among patients with head AIS scores > 1. A total of 996 patients were included, of whom 786 (78.9%) were males. In-hospital mortality, ICU admission, neurosurgical intervention, and prolonged total hospital length of stay (THLOS) were recorded for 27 (2.7%), 207 (20.8%), 82 (8.2%), and 205 (20.6%) patients, respectively. For predicting in-hospital mortality, all scoring systems had AUROC point estimates above 0.9 and 0.75 among all included patients and patients with head AIS > 1, respectively, without any significant differences. The Marshall and NIRIS scoring systems had higher AUROCs for predicting ICU admission and neurosurgery than the other scoring systems. For predicting THLOS ≥ seven days, although the NIRIS and Marshall scoring systems seemed to have higher AUROC point estimates when all patients were analyzed, five scoring systems performed roughly the same in the head AIS > 1 subgroup.

Confirmatory digital subtraction angiography after clinical brain death/death by neurological criteria: impact on number of donors and organ transplants.

Demand for organs exceeds the number of transplants available, underscoring the need to optimize organ donation procedures. However, protocols for determining brain death (BD)/death by neurological criteria (DNC) vary considerably worldwide. In Denmark, digital subtraction angiography (DSA) is the only legally approved confirmatory test for diagnosing BD/DNC. We investigated the effect of the time delay caused by (repeat) confirmatory DSA on the number of organs donated by patients meeting clinical criteria for BD/DNC. We hypothesized that, first, patients investigated with ≥2 DSAs donate fewer organs than those investigated with a single DSA; second, radiological interpretation of DSA is subject to interrater variability; and third, residual intracranial circulation is inversely correlated with inotropic blood pressure support. All DSAs performed over a 7-year period as part of BD/DNC protocols at Rigshospitalet, Copenhagen University Hospital, Denmark, were included. Clinical data were extracted from electronic health records. DSAs were reinterpreted by an independent neurinterventionist blinded to the original radiological reports. We identified 130 DSAs in 100 eligible patients. Patients with ≥2 DSAs (n = 20) donated fewer organs (1.7 +/- 1.6 SD) than patients undergoing a single DSA (n = 80, 2.6 +/- 1.7 organs, p = 0.03), and they became less often donors (n = 12, 60%) than patients with just 1 DSA (n = 65, 81.3%; p = 0.04). Interrater agreement of radiological DSA interpretation was 88.5% (Cohen's kappa = 0.76). Patients with self-maintained blood pressure had more often residual intracranial circulation (n = 13/26, 50%) than patients requiring inotropic support (n = 14/74, 18.9%; OR = 0.23, 95% CI [0.09-0.61]; p = 0.002). In potential donors who fulfill clinical BD/DNC criteria, delays caused by repetition of confirmatory DSA result in lost donors and organ transplants. Self-maintained blood pressure at the time of clinical BD/DNC increases the odds for residual intracranial circulation, creating diagnostic uncertainty because radiological DSA interpretation is not uniform. We suggest that avoiding unnecessary repetition of confirmatory investigations like DSA may result in more organs donated.

Artificial intelligence-based computer-assisted detection/diagnosis (AI-CAD) for screening mammography: Outcomes of AI-CAD in the mammographic interpretation workflow

PurposeTo evaluate the stand-alone diagnostic performances of AI-CAD and outcomes of AI-CAD detected abnormalities when applied to the mammographic interpretation workflow. MethodsFrom January 2016 to December 2017, 6499 screening mammograms of 5228 women were collected from a single screening facility. Historic reads of three radiologists were used as radiologist interpretation. A commercially-available AI-CAD was used for analysis. One radiologist not involved in interpretation had retrospectively reviewed the abnormality features and assessed the significance (negligible vs. need recall) of the AI-CAD marks. Ground truth in terms of cancer, benign or absence of abnormality was confirmed according to histopathologic diagnosis or negative results on the next-round screen. ResultsOf the 6499 mammograms, 6282 (96.7%) were in the negative, 189 (2.9%) were in the benign, and 28 (0.4%) were in the cancer group. AI-CAD detected 5 (17.9%, 5 of 28) of the 9 cancers that were intially interpreted as negative. Of the 648 AI-CAD recalls, 89.0% (577 of 648) were marks seen on examinations in the negative group, and 267 (41.2%) of the AI-CAD marks were considered to be negligible. Stand-alone AI-CAD has significantly higher recall rates (10.0% vs. 3.4%, P < 0.001) with comparable sensitivity and cancer detection rates (P = 0.086 and 0.102, respectively) when compared to the radiologists’ interpretation. ConclusionAI-CAD detected 17.9% additional cancers on screening mammography that were initially overlooked by the radiologists. In spite of the additional cancer detection, AI-CAD had significantly higher recall rates in the clinical workflow, in which 89.0% of AI-CAD marks are on negative mammograms.

Postoperative Chest X-rays After Open Reduction Internal Fixation of Clavicle Fractures May Not Be Cost-effective or Necessary.

To assess the utility of chest x-rays following open reduction internal fixation of clavicle fractures. Particularly in terms of detection of acute postoperative pneumothorax and cost- effectiveness of obtaining routine chest x-rays post-operatively. A retrospective cohort study. Level I trauma centerPatients 236 patients who underwent ORIF from 2013 to 2020 between the ages of 12 and 93. Chest x-ray performed post-operatively. Presence of acute postoperative pneumothorax. Of the 236 patients who underwent surgery 189 (80%) of patients received a CXR post-operatively 7 (3%) of patients experienced respiratory symptoms. All patients who had respiratory symptoms received a post-operative CXR. Those who did not receive a CXR post-operatively did not experience any respiratory complications. Two patients in the cohort had a postoperative pneumothorax, which were both present pre-operatively and unchanged in size post-operatively. Both of these patients were managed with general anesthesia and endotracheal intubation for surgery. The most common finding on CXR post-operatively was atelectasis. The cost of a portable CXR can be upwards of $594 when including technology, personnel fees, and radiological interpretation. Post-operative chest x-rays after clavicle open reduction and internal fixation did not detect any acute postoperative pneumothorax in asymptomatic patients. It is not cost-effective to routinely get chest x-rays in patients following open reduction internal fixation of clavicle fractures. In our study, of the 189 chest x-rays performed, only seven patients experienced postoperative respiratory symptoms. Our healthcare system as a total could have saved upwards of $108,108 in total for these patients as they may have been considered non-reimbursable by an insurance provider.

The trend in Management of Stable Burst Fracture of Thoracolumbar Spine with Intact Neurology among Members of the Association of Spine Surgeons of Nepal: A Survey

Background:The management of stable burst fracture of thoracolumbar spine with intact neurology is controversial. The radiological interpretation of such fractures differs among the surgeons and hence the treatment. So, this survey aims to study discrepancies in management of such fractures among the members of the Association of Spine Surgeons of Nepal (ASSN). Methods: Two representative cases of stable burst fracture with intact neurology (Case 1 AO type A3 and Case 2 AO type A4) were chosen by the authors. Questionnaire was formulated in Google form which also contained X-rays and CT scans of the cases. Google form was circulated among the members of ASSN. The response was recorded in excel and was analyzed. Results:Out of 67 members of ASSN, 55 (82.08%) responded the survey. 35 responders (63.63%) used both AO classification and TLICS. Majority of responders responded correctly as AO Type A3 and A4 burst fractures after looking at X-rays and CT scans of Case 1 and Case 2 respectively. Regardless of TLICS of 0-3 or 4 in both the cases, majority (81.81% in Case 1 and 78.18% in Case 2) chose surgical intervention. The degree of kyphosis for 48 respondents (87.27%), the level of fracture for 45 respondents (81.81%), bone quality for 41 respondents (74.54%), the severity of canal compromise for 40 respondents (72.72%) were some important factors in treatment decisions. Conclusion:The wide interobserver variability exists among members of ASSN in the management of stable burst fractures of the thoracolumbar spine with intact neurology. KeyWords: Intact neurology, Interobserver variability, Stable burst fracture, Trend in management

An Apparent Diffusion Coefficient-Based Machine Learning Model Can Improve Prostate Cancer Detection in the Grey Area of the Prostate Imaging Reporting and Data System Category 3: A Single-Centre Experience

The Prostate Imaging and Reporting Data System (PI-RADS) has a key role in the management of prostate cancer (PCa). However, the clinical interpretation of PI-RADS 3 score lesions may be challenging and misleading, thus postponing PCa diagnosis to biopsy outcome. Multiparametric magnetic resonance imaging (mpMRI) radiomic analysis may represent a stand-alone noninvasive tool for PCa diagnosis. Hence, this study aims at developing a mpMRI-based radiomic PCa diagnostic model in a cohort of PI-RADS 3 lesions. We enrolled 133 patients with 155 PI-RADS 3 lesions, 84 of which had PCa confirmation by fusion biopsy. Local radiomic features were generated from apparent diffusion coefficient maps, and the four most informative were selected using LASSO, the Wilcoxon rank-sum test (p &lt; 0.001), and support vector machines (SVMs). The selected features where augmented and used to train an SVM classifier, externally validated on a holdout subset. Linear and second-order polynomial kernels were exploited, and their predictive performance compared through receiver operating characteristics (ROC)-related metrics. On the test set, the highest performance, equally for both kernels, was specificity = 76%, sensitivity = 78%, positive predictive value = 80%, and negative predictive value = 74%. Our findings substantially improve radiologist interpretation of PI-RADS 3 lesions and let us advance towards an image-driven PCa diagnosis.

Lista de verificación: tomografía computarizada de cuello en urgencias no traumáticas

Patients attending the emergency department (ED) with cervical inflammatory/infectious symptoms or presenting masses that may involve the aerodigestive tract or vascular structures require a contrast-enhanced computed tomography (CT) scan of the neck. Its radiological interpretation is hampered by the anatomical complexity and pathophysiological interrelationship between the different component systems in a relatively small area.Recent studies propose a systematic evaluation of the cervical structures, using a 7-item checklist, to correctly identify the pathology and detect incidental findings that may interfere with patient management.As a conclusion, the aim of this paper is to review CT findings in non-traumatic pathology of the neck in the ED, highlighting the importance of a systematic approach in its interpretation and synthesis of a structured, complete, and concise radiological report.

Clinical and imaging modality factors impacting radiological interpretation of breast screening in young women with neurofibromatosis type 1.

Young women with Neurofibromatosis type 1 (NF1) have a high risk of developing breast cancer and poorer survival following breast cancer diagnosis. International guidelines recommend commencing breast screening between 30 and 35 years; however, the optimal screening modality is unestablished, and previous reports suggest that breast imaging may be complicated by the presence of intramammary and cutaneous neurofibromas (cNFs). The aim of this study was to explore potential barriers to implementation of breast screening for young women with NF1.Twenty-seven women (30-47 years) with NF1 completed breast screening with breast MRI, mammogram and breast ultrasound. Nineteen probably benign/suspicious lesions were detected across 14 women. Despite the presence of breast cNFs, initial biopsy rate for participants with NF1 (37%), were comparable to a BRCA pathogenic variant (PV) cohort (25%) (P = 0.311). No cancers or intramammary neurofibromas were identified. Most participants (89%) returned for second round screening.The presence of cNF did not affect clinician confidence in 3D mammogram interpretation, although increasing breast density, frequently seen in young women, impeded confidence for 2D and 3D mammogram. Moderate or marked background parenchymal enhancement on MRI was higher in the NF1 cohort (70.4%) than BRCA PV carriers (47.3%), which is an independent risk factor for breast cancer.Breast MRI was the preferred mode of screening over mammogram, as the majority (85%) with NF1 demonstrated breast density (BI-RADS 3C/4D), which hinders mammogram interpretation. For those with high breast density and high cNF breast coverage, 3D rather than 2D mammogram is preferred, if MRI is unavailable.

Foreign body in the external auditory canal described as a pseudotumor of the middle ear: a case report

&amp;lt;p&amp;gt;&amp;lt;strong&amp;gt;Introduction. &amp;lt;/strong&amp;gt;The aim of this article was to present a case of a foreign body in the external auditory canal described as a pseudotumor of the middle ear, as well as to point out diagnostic and therapeutic aspects of this problem.&amp;lt;/p&amp;gt; &amp;lt;p&amp;gt;&amp;lt;strong&amp;gt;Case report. &amp;lt;/strong&amp;gt;An 8-year-old girl was hospitalized several times in our department due to the surgery of left-sided chronic otitis media. Mastoidectomy and posterior atticotomy were performed during initial hospitalization. Six months later, she was admitted for the second act of the left-sided tympanoplasty. However, two months prior hospitalization, symptoms regarding the right ear appeared: sense of fullness, gradual hearing loss and occasional pain. An otoscopic finding on the right indicated the presence of &amp;amp;ldquo;tumefaction&amp;amp;rdquo; in the external auditory canal with a surface that was markedly hyperemic. This pseudotumor was finally interpreted as the foreign body in the external auditory canal.&amp;lt;/p&amp;gt; &amp;lt;p&amp;gt;&amp;lt;strong&amp;gt;Conclusion. &amp;lt;/strong&amp;gt;Foreign bodies in the external auditory canal (EAC) are relatively common in pediatric population. The medical history sometimes is not reliable, and the clinical and radiological interpretation of pseudotumor in the external auditory canal or middle ear must include this possibility in the differential diagnosis as well. Surgical exploration and pathohistological diagnosis are necessary to make a definitive diagnosis and to avoid potential complications.&amp;lt;/p&amp;gt;

IMG-08. VOLUMETRIC MEASUREMENT OF TUMOR SIZE OUTPERFORMS STANDARD TWO-DIMENSIONAL METHOD IN EARLY PREDICTION OF TUMOR PROGRESSION IN PEDIATRIC GLIOMA

Abstract BACKGROUND Pediatric brain tumors (PBTs) are the leading cause of cancer-related death in children. Currently, based on the RAPNO guidelines, two-dimensional (2D) measurements are used to detect progression. However, due to the complex radiographic appearance of many PBTs, 2D measurements may not accurately reflect tumor growth. Here, we compared 3D, 2D, and qualitative radiologists’ interpretations in a group of PBTs to determine tumor progression. METHODS Six PBT patients (5 low-grade glioma, 1 ependymoma, age range at the time of baseline scan: 23-191 months) with an average of 5 imaging time points who had at least one episode of progression with subsequent surgery were included. Segmentation was performed on either T1 post-contrast or FLAIR, whichever that best identified the tumor. For 2D measurements, the two largest perpendicular diameters in a section that included the largest tumor component were measured. For the volumetric assessment, the tumors were manually segmented, and the volume was computed using the ITK-SNAP software. In multifocal tumors, only the tumor component that showed progression and underwent resection was included. A 25% and 41% increase in tumor size was considered as progression on 2D and 3D measurements, respectively. The time to progression (TTP) based on 2D vs 3D assessment as well as the official pediatric neuroradiologist interpretation based on the electronic health record report was compared. RESULTS In three of six patients, volumetric segmentation detected tumor progression at an earlier time point compared to 2D measurements and radiologists’ interpretation (median TTP: 134 compared to 503 for 2D and radiologist interpretation). For these three patients, the TTP based on 3D vs 2D assessment was: 252 vs 1299 days, 134 vs 503 days, and 80 vs 255 days. CONCLUSION Volumetric measurements can determine tumor progression earlier than the current standard method of 2D measurements and qualitative interpretation of radiologists in PBTs.

Viral Detection Is Associated With Severe Disease in Children With Suspected Community-Acquired Pneumonia.

To evaluate the role of virus detection on disease severity among children presenting to the emergency department (ED) with suspected community-acquired pneumonia (CAP). We performed a single-center prospective study of children presenting to a pediatric ED with signs and symptoms of a lower respiratory tract infection and who had a chest radiograph performed for suspected CAP. We included patients who had virus testing, with results classified as negative for virus, human rhinovirus, respiratory syncytial virus (RSV), influenza, and other viruses. We evaluated the association between virus detection and disease severity using a 4-tiered measure of disease severity based on clinical outcomes, ranging from mild ( discharged from the ED) to severe (receipt of positive-pressure ventilation, vasopressors, thoracostomy tube placement, or extracorporeal membrane oxygenation, intensive care unit admission, diagnosis of severe sepsis or septic shock, or death) in models adjusted for age, procalcitonin, C-reactive protein, radiologist interpretation of the chest radiograph, presence of wheeze, fever, and provision of antibiotics. Five hundred seventy-three patients were enrolled in the parent study, of whom viruses were detected in 344 (60%), including 159 (28%) human rhinovirus, 114 (20%) RSV, and 34 (6%) with influenza. In multivariable models, viral infections were associated with increasing disease severity, with the greatest effect noted with RSV (adjusted odds ratio [aOR], 2.50; 95% confidence interval [CI], 1.30-4.81) followed by rhinovirus (aOR, 2.18; 95% CI, 1.27-3.76). Viral detection was not associated with increased severity among patients with radiographic pneumonia (n = 223; OR, 1.82; 95% CI, 0.87-3.87) but was associated with severity among patients without radiographic pneumonia (n = 141; OR, 2.51; 95% CI, 1.40-4.59). The detection of a virus in the nasopharynx was associated with more severe disease compared with no virus; this finding persisted after adjustment for age, biomarkers, and radiographic findings. Viral testing may assist with risk stratification of patients with lower respiratory tract infections.

Validation of a deep learning model for traumatic brain injury detection and NIRIS grading on non-contrast CT: a multi-reader study with promising results and opportunities for improvement.

This study aimed to assess and externally validate the performance of a deep learning (DL) model for the interpretation of non-contrast computed tomography (NCCT) scans of patients with suspicion of traumatic brain injury (TBI). This retrospective and multi-reader study included patients with TBI suspicion who were transported to the emergency department and underwent NCCT scans. Eight reviewers, with varying levels of training and experience (two neuroradiology attendings, two neuroradiology fellows, two neuroradiology residents, one neurosurgery attending, and one neurosurgery resident), independently evaluated NCCT head scans. The same scans were evaluated using the version 5.0 of the DL model icobrain tbi. The establishment of the ground truth involved a thorough assessment of all accessible clinical and laboratory data, as well as follow-up imaging studies, including NCCT and magnetic resonance imaging, as a consensus amongst the study reviewers. The outcomes of interest included neuroimaging radiological interpretation system (NIRIS) scores, the presence of midline shift, mass effect, hemorrhagic lesions, hydrocephalus, and severe hydrocephalus, as well as measurements of midline shift and volumes of hemorrhagic lesions. Comparisons using weighted Cohen's kappa coefficient were made. The McNemar test was used to compare the diagnostic performance. Bland-Altman plots were used to compare measurements. One hundred patients were included, with the DL model successfully categorizing 77 scans. The median age for the total group was 48, with the omitted group having a median age of 44.5 and the included group having a median age of 48. The DL model demonstrated moderate agreement with the ground truth, trainees, and attendings. With the DL model's assistance, trainees' agreement with the ground truth improved. The DL model showed high specificity (0.88) and positive predictive value (0.96) in classifying NIRIS scores as 0-2 or 3-4. Trainees and attendings had the highest accuracy (0.95). The DL model's performance in classifying various TBI CT imaging common data elements was comparable to that of trainees and attendings. The average difference for the DL model in quantifying the volume of hemorrhagic lesions was 6.0mL with a wide 95% confidence interval (CI) of - 68.32 to 80.22, and for midline shift, the average difference was 1.4mm with a 95% CI of - 3.4 to 6.2. While the DL model outperformed trainees in some aspects, attendings' assessments remained superior in most instances. Using the DL model as an assistive tool benefited trainees, improving their NIRIS score agreement with the ground truth. Although the DL model showed high potential in classifying some TBI CT imaging common data elements, further refinement and optimization are necessary to enhance its clinical utility.