Supine Abdominal Radiographs Research Articles

Deep learning-based fully automatic Risser stage assessment model using abdominal radiographs.

Artificial intelligence has been increasingly used in medical imaging and has demonstrated expert level performance in image classification tasks. To develop a fully automatic approach for determining the Risser stage using deep learning on abdominal radiographs. In this multicenter study, 1,681 supine abdominal radiographs (age range, 9-18years, 50% female) obtained between January 2019 and April 2022 were collected retrospectively from three medical institutions and graded manually using the United States Risser staging system. A total of 1,577 images from Hospitals 1 and 2 were used for development, and 104 images from Hospital 3 for external validation. From each radiograph, right and left iliac crest patch images were extracted using the pelvic bone segmentation model DeepLabv3 + with the EfficientNet-B0 encoder trained with 90 digitally reconstructed radiographs from pelvic computed tomography scans with a pelvic bone mask. Using these patch images, ConvNeXt-B was trained to grade according to the Risser classification. The model's performance was evaluated using accuracy, area under the receiver operating characteristic curve (AUROC), and mean absolute error. The fully automatic Risser stage assessment model showed an accuracy of 0.87 and 0.75, mean absolute error of 0.13 and 0.26, and AUROC of 0.99 and 0.95 on internal and external test sets, respectively. We developed a deep learning-based, fully automatic segmentation and classification model for Risser stage assessment using abdominal radiographs.

Deep Learning-Enabled Detection of Pneumoperitoneum in Supine and Erect Abdominal Radiography: Modeling Using Transfer Learning and Semi-Supervised Learning.

Detection of pneumoperitoneum using abdominal radiography, particularly in the supine position, is often challenging. This study aimed to develop and externally validate a deep learning model for the detection of pneumoperitoneum using supine and erect abdominal radiography. A model that can utilize "pneumoperitoneum" and "non-pneumoperitoneum" classes was developed through knowledge distillation. To train the proposed model with limited training data and weak labels, it was trained using a recently proposed semi-supervised learning method called distillation for self-supervised and self-train learning (DISTL), which leverages the Vision Transformer. The proposed model was first pre-trained with chest radiographs to utilize common knowledge between modalities, fine-tuned, and self-trained on labeled and unlabeled abdominal radiographs. The proposed model was trained using data from supine and erect abdominal radiographs. In total, 191212 chest radiographs (CheXpert data) were used for pre-training, and 5518 labeled and 16671 unlabeled abdominal radiographs were used for fine-tuning and self-supervised learning, respectively. The proposed model was internally validated on 389 abdominal radiographs and externally validated on 475 and 798 abdominal radiographs from the two institutions. We evaluated the performance in diagnosing pneumoperitoneum using the area under the receiver operating characteristic curve (AUC) and compared it with that of radiologists. In the internal validation, the proposed model had an AUC, sensitivity, and specificity of 0.881, 85.4%, and 73.3% and 0.968, 91.1, and 95.0 for supine and erect positions, respectively. In the external validation at the two institutions, the AUCs were 0.835 and 0.852 for the supine position and 0.909 and 0.944 for the erect position. In the reader study, the readers' performances improved with the assistance of the proposed model. The proposed model trained with the DISTL method can accurately detect pneumoperitoneum on abdominal radiography in both the supine and erect positions.

Clinics in diagnostic imaging (216).

Clinics in diagnostic imaging (216).

Supine and erect abdominal radiographs are both feasible in patients with acute abdomen.

DEAR EDITOR, We have read the article entitled “Diagnostic values of supine and erect abdominal radiographs for patients with acute abdomen: which is better for decision making?” published in the June issue of the Journal of the Chinese Medical Association with interest.1 Lee et al.1 tried to evaluate whether supine or erect abdominal radiographs are more suitable in cases of acute abdomen. The authors found that there was no significant difference in diagnostic sensitivity between the two modalities among all causes-inducing acute abdomen. As for bowel obstruction, the study revealed statistically better diagnostic relevance in erect abdominal radiographs. We congratulated the authors’ successful publication but there are some questions that need clarification. First, is it possible that one computed tomography (CT) scan reported more than one diagnosis? For example, a patient might simultaneously have a tubo-ovarian abscess associated with either adynamic ileus or obstructive bowel syndrome. Could the authors kindly provide the rationale about their strategy to evaluate the acute abdomen or whether their study would influence the further therapeutic plan (surgical intervention or conservative treatment)? Second, the authors concluded that erect abdominal radiographs were strongly suggested to use if highly suspected bowel obstruction. We appreciated the authors’ effort to propose this recommendation; however, no matter whether an erect or supine abdominal radiograph was used, it may hardly influence the decision and the subsequent management as shown above. In cases with extremely severe abdominal pain or uncertain diagnosis of acute abdomen, most of the clinicians may arrange a CT scan for further evaluation regardless of abdominal radiograph findings.2–5 Besides, it is uncomfortable and difficult for patients with acute abdomen to keep an erect posture. Additionally, the position of standing may be much detorted during the examination because of intolerance of pain. Despite the aforementioned questions, the authors provided evidence of abdominal radiographs choice regarding to acute abdomen.1 Based on this study, clinicians may save their time and effort to struggle with whether erect or supine abdominal radiographs should be taken. Though the CT scan had relatively higher accuracy, the importance and role of conventional radiographs or much convenient and little or absence of radiation-exposure ultrasound in the setting of emergency department should not be neglected.6,7 Especially in the era of the coronavirus disease 2019 (COVID-19) pandemic8,9, timely CT scan may be less available to individuals with acute abdomen. Also, conventional radiographs bring the least pressure to staff in hospital than other modalities for fear of getting infection.10 Overall, we appreciate the authors’ great work focusing on this topic. We hope to learn more from the authors with positive response. ACKNOWLEDGMENTS This article was supported by grants from the Taiwan Ministry of Science and Technology, Executive Yuan, Taiwan (MOST 110-2314-B-075-016-MY3 and MOST 111-2314-B-075-045), and Taipei Veterans General Hospital (V110C-082, and V111C-103). The authors appreciate the support from Female Cancer Foundation, Taipei, Taiwan.

Diagnostic values of supine and erect abdominal radiographs for patients with acute abdomen: which is better for decision making?

Plain abdominal radiography including supine and erect abdominal radiographs (SAR and EAR) is a frequently used image modality for preliminary evaluation of acute abdomen. We aimed to explore which one of the SAR or EAR has a higher diagnostic value in overall diagnosis of acute abdomen, including their respective advantages over each other for the various underlying diseases. We retrospectively analyzed the imaging findings of plain abdominal radiography of patients with acute abdomen who received abdominal computed tomography (CT) within 24 hours at the emergency department of a medical center in northern Taiwan between October 2019 and February 2020. Final diagnosis was made by CT reports and clinical data. The relevance between the imaging findings and clinical diagnosis in the groups of SAR and EAR were compared. A total of 1009 cases with acute abdomen were included, of which 341 (33.8%) underwent EAR and 668 (66.2%) underwent SAR. Among them, 820 cases had final diagnosis confirmed by CT and clinical data. In comparison of the diagnostic relevance of SAR and EAR, there were no significant difference in the overall acute abdomen, but EAR showed a better diagnostic relevance in cases with bowel obstruction than SAR did (100% vs 87.2%, p < 0.05). No statistical difference in other abdominal diseases. There is no significant difference between SAR and EAR in evaluation of overall acute abdomen. However, EARs has a diagnostic advantage over SAR for evaluation of suspected bowel obstruction.

Çocuklarda Hirschsprung Hastalığı Cerrahi Tedavisi Sonuçları: 12 Yıllık Bir Deneyim

Objective: In the treatment of Hirschsprung’s disease, many surgical techniques have been developed and modified including laparotomy, resection of the affected colon segment, and pull-through of the functional bowel segment to the anus. Conventional surgical methods have many early and late postoperative complications, albeit at a reduced rate today. In this study, it was aimed to compare the treatment results of children with Hirschsprung Disease who were operated with single or two-stage techniques in our department. Method: Hospital records of the patients operated for Hirschsprung disease were retrospectively evaluated. Demografic characteristics, findings of physical and rectal examinations, X-rays findings, rectal biopsy results, surgical treatment methods and outcomes were analyzed. Results: Fifty-two children including 41-male (78.8%) and 11-female (21.2%) patients with age ranging from 1-day to 5-year and mean age of 6.8-day in neonates and 6.3-month in others were enrolled in the study. All patients presented with acute or subacute intestinal obstruction. Levels were observed on supine abdominal radiographs in all; transition zone and retention of opaque material were observed on opaque enema or retention radiographs.In patients whose rectal biopsies revealed aganglionic megacolon two staged procedures were performed including initial colostomy followed by definitive procedure of Duhamel with stapler in 30, and Soave in 12, and Swenson in 1 patient. Nine patients were operated with one-stage endorectal pull-through technique. The complication rate in 43 patients with Hirschsprung disease managed by two staged operative procedure was (78.8%), and it was 11.1% in 9 cases that underwent transanal endorectal pullthrough procedure. It was learned that effective anal dilatation was not performed in 22-patients who had postoperative enterocolitis attacks. Conclusion: In addition to the low complication rate, the morbidity rates in the transanal endorectal pull-through method, which achieved successful results with only one procedure, were very low compared to multiple-stage operations.

The value of the erect abdominal radiograph for the diagnosis of mechanical bowel obstruction and paralytic ileus in adults presenting with acute abdominal pain.

IntroductionThere is discord on the value of the erect abdominal radiograph for diagnosing acute abdominal pathologies. The erect radiograph can be uncomfortable for patients in pain and increases patient radiation dose.AimTo determine if including the erect abdominal radiograph in plain abdominal radiography (PAR) improved diagnostic accuracy for identifying mechanical bowel obstruction and/or paralytic ileus in adults presenting with acute abdominal pain.Methods PAR of 40 consecutive adults presenting with suspected bowel obstruction or paralytic ileus was retrospectively sampled and independently reviewed by two emergency department (ED) consultants and two radiology consultants for bowel obstruction and paralytic ileus across two sessions. In session 1, the assessors assessed the supine abdominal radiographs (PAR 1) and clinical details in a randomised order, and session 2, at least 6 weeks later, they assessed the supine and erect radiographs (PAR 2) and clinical details of the randomly re‐ordered cases. Computed tomography was the reference standard. Pair‐wise comparisons of receiver operating characteristic curves were calculated to assess for significant differences in participants’ diagnostic accuracy using MedCalc 16.4.3.ResultsAverage sensitivity, specificity and area under the receiver operating characteristic curves (AUROC) were 69.7%, 61.0% and 0.642 for PAR 1, respectively, and 80.0%, 53.4% and 0.632 for PAR 2 respectively. For AUROC there were no significant differences (P > 0.05) between PAR 1 and PAR 2. Intra‐rater and inter‐rater agreement improved in PAR 2.ConclusionThere was no statistically significant improvement in diagnostic accuracy when including the erect radiograph in PAR for the acute abdomen.

Detection of high-grade small bowel obstruction on conventional radiography with convolutional neural networks.

The purpose of this pilot study is to determine whether a deep convolutional neural network can be trained with limited image data to detect high-grade small bowel obstruction patterns on supine abdominal radiographs. Grayscale images from 3663 clinical supine abdominal radiographs were categorized into obstructive and non-obstructive categories independently by three abdominal radiologists, and the majority classification was used as ground truth; 74 images were found to be consistent with small bowel obstruction. Images were rescaled and randomized, with 2210 images constituting the training set (39 with small bowel obstruction) and 1453 images constituting the test set (35 with small bowel obstruction). Weight parameters for the final classification layer of the Inception v3 convolutional neural network, previously trained on the 2014 Large Scale Visual Recognition Challenge dataset, were retrained on the training set. After training, the neural network achieved an AUC of 0.84 on the test set (95% CI 0.78-0.89). At the maximum Youden index (sensitivity+specificity-1), the sensitivity of the system for small bowel obstruction is 83.8%, with a specificity of 68.1%. The results demonstrate that transfer learning with convolutional neural networks, even with limited training data, may be used to train a detector for high-grade small bowel obstruction gas patterns on supine radiographs.

Inverted V sign.

Despite the ever-increasing utilization of CT, a supine abdominal radiograph can still be the ‘‘presenting’’ imaging study for patients with acute abdominal pain. Although less sensitive than upright or left lateral decubitus views, a supine frontal abdomen view thus can present the initial opportunity to detect pneumoperitoneum [1]. Among the many described radiographic signs is an ‘‘inverted V’’ visible in the pelvis (Fig. 1). Superomedially directed oblique lines are thought to represent gas outlining peritoneal folds that overlie either the obliterated umbilical arteries [2], or the inferior epigastric arteries [3]. As with any imaging evidence of pneumoperitoneum, the inverted V should prompt a search for perforated viscus.

Spectrum of Signs of Pneumoperitoneum

Pneumoperitoneum is caused by rupture of a hollow viscus that includes the stomach, small bowel, and large bowel, with the exception of those portions that are retroperitoneal in the duodenum and colon. The causes of pneumoperitoneum are numerous, ranging from iatrogenic and benign causes to more life-threatening conditions. In the absence of a benign cause of pneumoperitoneum, the identification of free intraperitoneal gas usually indicates the need for emergency surgery to repair a perforated bowel. The plain film is the primary diagnostic tool for detecting pneumoperitoneum: multiple signs of free intraperitoneal air can be found especially on supine abdominal radiographs. Computed tomography (CT) examination has been shown to be more sensitive than abdominal radiographs for the detection of free intraperitoneal air. It is important that the radiologist become familiar with the signs of pneumoperitoneum that can be discerned on abdominal radiographs, on CT scout view, and on CT scan.

An Unusual Cause of Anterior Abdominal Wall Abscess

Question: A 71-year-old woman, who had no previous psychiatric history or abdominal surgery, was admitted to the hospital for left hypochondrial mass for 2 weeks, associated with an ache over the area and loss of appetite for the same duration. No fever was experienced. She was also noted to move her bowels every 2–3 days (previously daily) for the past 2 weeks. Physical examination revealed a 7-cm, immobile, firm mass about 6 cm below the left subcostal margin with normal overlying skin. Digital rectal examination was unremarkable. Initial abnormal laboratory tests were as follows: hemoglobin, 11.0g/dL; white blood cell count, 11.6 10/L; and sodium, 127 mmol/L. Supine and erect abdominal radiographs showed gaseous distension of the large bowel (Figure A). CT of the abdomen and pelvis was done (Figure B, C). What is your diagnosis? See the Gastroenterology web site (www.gastrojournal.org) for more information on submitting your favorite image to Clinical Challenges and Images in GI.

An evaluation of embolisation in the management of haemorrhage from traumatic solid organ injury at a level 1 major trauma centre

An evaluation of embolisation in the management of haemorrhage from traumatic solid organ injury at a level 1 major trauma centre

Rigler’s sign and the football sign

Rigler’s sign was first described in 1941 by L G Rigler as a new radiological sign for recognising free air in the peritoneal cavity on supine radiograph. The presence of pneumoperitoneum allows free intraperitoneal air to be contrasted with intraluminal gas, accentuating the wall of gas-containing viscera. It is observed in infants and very ill patients where only limited radiographs of the abdomen are possible. The football sign was first described by R E Miller in the 1960s. Seen on supine abdominal radiographs, this describes an oval radiolucency resembling an American football. It is important for the radiologist to recognise the supporting signs of pneumoperitoneum, such as Rigler’s sign and the football sign, on supine abdominal radiographs, especially in neonates and infants, where erect chest/abdominal radiographs are not always possible.

Umbilical venous catheter position: evaluation by ultrasound

Umbilical venous catheter (UVC) insertion is frequently performed in critically ill neonates, with catheter position evaluated by an abdominal radiograph. The position of the catheter can be difficult to assess based on supine film alone. We aimed to determine whether ultrasound enables precise evaluation of the catheter tip position. We prospectively evaluated UVC position in 75 neonates by ultrasound and supine abdominal radiograph. Distance from the catheter tip to the diaphragm was measured for both modalities. Anatomical location of the catheter was recorded for both studies. The Student's t-test was used to compare the means of the differences in catheter position on radiograph and ultrasound. In 46 patients there was no difference in the distance between the catheter tip and the diaphragm on ultrasound and radiograph. In 29 babies there was a difference of 1-to-7 mm. Ultrasound identified malposition of the catheter in three cases in which the position of the catheter on the plain film was deemed acceptable. Our study suggests that ultrasound is reliable and may replace plain radiograph for the identification of the UVC position, preventing complications from malposition, and reducing exposure to ionizing radiation.

Relationships Among the Lactulose Breath Test, Intestinal Gas Volume, and Gastrointestinal Symptoms in Patients with Irritable Bowel Syndrome

Irritable bowel syndrome (IBS) patients frequently complain of gas-related symptoms, and the lactulose breath test (LBT) is a test that assesses the amount of fermented gas generated by bacteria in the bowel. We aimed to assess the relationship among intestinal gas volume, LBT result, and gastrointestinal symptom score in healthy control and functional bowel disorder (FBD) subjects. In 84 IBS subjects, 24 FBD subjects other than IBS, and 25 healthy controls, a symptom questionnaire that enquired about seven main symptoms, plain abdominal radiography and a LBT were checked on the same day. The intestinal gas volume was calculated as the gas volume score (GVS) with a digitalized image of plain supine abdominal radiographs. The GVS was greater in the LBT (+) group compared to the LBT (-) group (P=0.02). The GVS was greater in the FBD and IBS groups than in the control group (P<0.01). The GVS showed low but positive correlations with the severity and frequency of bloating, flatulence, abdominal pain, constipation, and tenesmus (P<0.05). The severity of flatulence (P=0.02) and the frequency of bloating (P=0.02) in the LBT (+) group were significantly higher than those in the LBT (-) group. Subjects with positive LBT had more gas-related symptoms and greater gas volume scores. Gas-related symptoms, positive LBT and increased GVS were significantly associated to each other. These findings can broaden the understanding of the pathophysiologic mechanisms of gas-related symptoms in IBS.

Milk of Calcium in Abdomen

A 45-year-old woman had an asymptomatic abnormality on a screening abdominal radiograph. The radiopaque mass in her right upper abdomen was surrounded by numerous "pearls" and resembled an abalone on the supine abdominal radiograph. We advised an additional upright abdominal radiograph, which showed a calcium fluid level. We also clarified the location of the cystic lesion at the right floating kidney, which changed its location between the supine and upright positions. Computed tomography of the abdomen revealed a right renal cyst with a calcium-fluid interface owing to the milk of calcium. The patient was then followed up without additional investigation or the need for intervention.

The role of plain radiographs in patients with acute abdominal pain at the ED

Objective The purpose of this study was to evaluate the added value of plain radiographs on top of clinical assessment in unselected patients presenting with acute abdominal pain at the emergency department (ED). Methods In a multicenter prospective trial, patients with abdominal pain more than 2 hours and less than 5 days presented at the ED were evaluated clinically, and a diagnosis was made by the treating physician. Subsequently, all patients underwent supine abdominal and upright chest radiographs, after which the diagnosis was reassessed by the treating physician. A final (reference) diagnosis was assigned by an expert panel. The number of changes in the primary diagnosis, as well as the accuracy of these changes, was calculated. Changes in the level of confidence were evaluated for unchanged diagnoses. Results Between March 2005 and November 2006, 1021 patients, 55% female, mean age 47 years (range, 19-94 years), were included. In 117 of 1021 patients, the diagnosis changed after plain radiographs, and this change was correct in 39 patients (22% of changed diagnoses and 4% of total study population). Overall, the clinical diagnosis was correct in 502 (49%) patients. The diagnosis after evaluation of the radiographs was correct in 514 (50%) patients, a nonsignificant difference ( P = .14). In 65% of patients with unchanged diagnosis before and after plain radiography, the level of confidence of that diagnosis did not change either. Conclusion The added value of plain radiographs is too limited to advocate their routine use in the diagnostic workup of patients with acute abdominal pain, because few diagnoses change and the level of confidence were mostly not affected.

Cecal volvulus: classic imaging appearance of an uncommon pediatric entity

A 13-year-old boy presented with a 1-day history of coffeeground emesis. Pertinent medical history included trisomy 17, developmental delay, and chronic constipation. Supine abdominal radiograph (Fig. 1) demonstrated gaseous distention of malpositioned colon located in the left upper quadrant. Upper GI (not shown) revealed malrotation without midgut volvulus. Enema (Fig. 2) showed beaked cut-off of contrast column (arrow), typical of cecal volvulus, and residual contrast agent from UGI (C). Both malrotation and cecal volvulus were surgically confirmed.

Are Supine Chest and Abdominal Radiographs the Best Way to Confirm PICC Placement in Neonates?

Peripherally inserted central catheters (PICCs) are commonly used in NICUs. Although they have many benefits, they also have many potential complications. Confirming catheter tip position is essential to decreasing complications, but the best method to achieve confirmation is unclear. Literature review for studies that address line position confirmation to assist health care providers in evaluating the available research and to identify gaps in the literature. A literature search of four major databases followed by an ancestry approach was performed. Articles reviewed specifically discuss PICC lines and PICC line placement confirmation. Data on confirming PICC placement were lacking. Fluoroscopic placement is ideal, but cannot be done at the bedside and is costly. Supine chest radiograph is the most widely used method and is convenient, but when line tip position is unclear, contrast or ultrasound confirmation can be used. When PICC lines are placed in the saphenous vein, infants may benefit from supine and lateral abdominal radiographs to ensure placement in the inferior vena cava. More studies are needed to generalize findings. PICC line tips should be located in the superior vena cava or inferior vena cava close to the junction with the right atrium (0.5-1 cm outside of the cardiac chambers in premature infants and 1-2 cm outside of the cardiac chambers in larger infants). Arm position is very important when performing radiographs for placement because movement of the arm can cause migration of the catheter. There is also significant inter-observer variability when identifying line tip position.

Acute ovarian torsion in an infant: diagnostic clues on supine and decubitus abdominal radiographs confirmed by ultrasound

Acute ovarian torsion is an uncommon yet important diagnostic consideration in any female pediatric patient presenting with acute abdominal pain. A 2-month-old infant presented to the emergency department with a several-day history of constipation. Evaluation with plain radiographs demonstrated a subtle yet persistent soft tissue mass in the right pelvis. Follow-up ultrasound revealed characteristic findings for ovarian torsion and subsequent salpingo-oophorectomy was performed confirming the diagnosis. Acute ovarian torsion is an uncommon and frequently overlooked diagnosis in female infants presenting with gastrointestinal or urinary tract symptoms. Careful assessment of plain radiographic findings may aid in the further management of this difficult diagnosis in female infants. Ultrasound is recommended to confirm the diagnosis.