A Convolutional Neural Network Model for Detecting Sellar Floor Destruction of Pituitary Adenoma on Magnetic Resonance Imaging Scans.
Ontology highlight
ABSTRACT: Objective
Convolutional neural network (CNN) is designed for image classification and recognition with a multi-layer neural network. This study aimed to accurately assess sellar floor invasion (SFI) of pituitary adenoma (PA) using CNN.Methods
A total of 1413 coronal and sagittal magnetic resonance images were collected from 695 patients with PAs. The enrolled images were divided into the invasive group (n = 530) and the non-invasive group (n = 883) according to the surgical observation of SFI. Before model training, 100 images were randomly selected for the external testing set. The remaining 1313 cases were randomly divided into the training and validation sets at a ratio of 80:20 for model training. Finally, the testing set was imported to evaluate the model performance.Results
A CNN model with a 10-layer structure (6-layer convolution and 4-layer fully connected neural network) was constructed. After 1000 epoch of training, the model achieved high accuracy in identifying SFI (97.0 and 94.6% in the training and testing sets, respectively). The testing set presented excellent performance, with a model prediction accuracy of 96%, a sensitivity of 0.964, a specificity of 0.958, and an area under the receptor operator curve (AUC-ROC) value of 0.98. Four images in the testing set were misdiagnosed. Three images were misread with SFI (one with conchal type sphenoid sinus), and one image with a relatively intact sellar floor was not identified with SFI.Conclusion
This study highlights the potential of the CNN model for the efficient assessment of PA invasion.
SUBMITTER: Feng T
PROVIDER: S-EPMC9289618 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:PurposeTo evaluate the use of preoperative virtual Magnetic Resonance Elastography (vMRE) for patients undergoing transsphenoidal resection of pituitary adenomas (PA).MethodsTen patients (60.2 ± 19.6 years; 8 males) were prospectively examined with the vMRE-method prior to transsphenoidal surgery. vMRE-images, reflecting tissue stiffness were reconstructed. From these images, histograms as well as the mean stiffness values over the tumor body were extracted. Finally, vMRE-data was compared with the PA consistency at surgery blinded to vMRE.ResultsIn all patients, successful vMRE-examination was performed enabling evaluation of even small PAs. For tumors with homogenous tissue, the mean stiffness value increased with surgical consistency grading. For heterogenous tumors, however, the mean stiffness value did not consistently reflect the grading at surgery. On the other hand, the vMRE-images and histograms were found to be able to characterize the tumor heterogeneity and display focal regions of high stiffness that were found to affect the surgery outcome in these PAs. The vMRE-images and histograms showed great promise in characterizing the consistency at surgery for these PAs.ConclusionEvaluation of PA consistency in preparation for surgery seems to be feasible using the vMRE-method. Our findings also address the need for high resolution diagnostic methods that can non-invasively display focal regions of increased stiffness, as such regions may increase the difficulty of transsphenoidal PA-resection.
| S-EPMC8270838 | biostudies-literature
Project description:PurposeThis study aimed to investigate the value of magnetic resonance (MR) characteristics in differentiating the subtypes of growth hormone pituitary adenomas.Materials and methodsThe clinical and MR imaging data of 70 patients with growth hormone pituitary adenoma confirmed by surgery and pathology were retrospectively analyzed. The tumors were divided into dense granular (DG; 36 cases) and sparse granular subtypes (SG; 34 cases). The tumors' MR features were analyzed, including the mean and maximum diameters, T2 signal intensity, T2 relative signal intensity (rSI), homogeneity, enhancement degree, and invasiveness (Knosp grade). Mann-Whitney U test and χ2 test were used to analyze MR characteristics between the 2 groups. The independent predictors and predictive probabilities of tumor subtypes were obtained via a logistic regression model, and the efficacy was compared by receiver operating characteristic curve.ResultsThe mean and maximum diameters of growth hormone adenoma in DG and SG were 1.77 versus 2.45 and 1.95 versus 3.00 cm (median, P < 0.05), respectively. There was a significant difference between the 2 groups in T2 signal intensity and rSI (P values were 0.02 and 0.001, respectively). Most DG adenomas (86.1%) appeared as hypointense on T2 images, and 38.2% of SG adenomas were hyperintense. There was no significant difference in tumor homogeneity (P = 0.622). A significant difference was found in the Knosp grade between the 2 subtypes (P = 0.004). In addition, the enhancement degree of SG adenomas was significantly higher than that of DG adenomas (P = 0.001). Logistic regression analysis showed that high T2 rSI value and marked contrast enhancement were independent predictors of the 2 subtypes, and the odds ratios were 4.811 and 4.649, respectively. The multivariate logistic model obtained relatively high predicting efficacy, and the area under the curve, sensitivity, and specificity were 0.765, 0.882, and 0.500, respectively.ConclusionsThere are significant differences in tumor size, T2 signal intensity, T2 rSI, enhancement degree, and invasiveness between DG and SG adenomas. The logistic model based on the marked contrast enhancement and high T2 rSI value has an important value in predicting the subtype of growth hormone adenoma.
| S-EPMC8763249 | biostudies-literature
Project description:Glomus tumor is a rare mesenchymal tumor with an organ-like structure. Sellar glomus tumors are extremely rare with only six reported cases in the literature. Because of the lack of special clinical manifestations and imaging features, the disorder may be easily misdiagnosed as other sellar tumors, especially pituitary adenomas. Here, the present study showed a case of a 69-year-old male with hypopituitarism who was preliminarily misdiagnosed as non-functional pituitary adenoma.
| S-EPMC9114742 | biostudies-literature
Project description:OBJECT High-resolution PET (hrPET) performed using a high-resolution research tomograph is reported as having a resolution of 2 mm and could be used to detect corticotroph adenomas through uptake of(18)F-fluorodeoxyglucose ((18)F-FDG). To determine the sensitivity of this imaging modality, the authors compared(18)F-FDG hrPET and MRI detection of pituitary adenomas in Cushing disease (CD). METHODS Consecutive patients with CD who underwent preoperative(18)F-FDG hrPET and MRI (spin echo [SE] and spoiled gradient recalled [SPGR] sequences) were prospectively analyzed. Standardized uptake values (SUVs) were calculated from hrPET and were compared with MRI findings. Imaging findings were correlated to operative and histological findings. RESULTS Ten patients (7 females and 3 males) were included (mean age 30.8 ± 19.3 years; range 11-59 years). MRI revealed a pituitary adenoma in 4 patients (40% of patients) on SE and 7 patients (70%) on SPGR sequences.(18)F-FDG hrPET demonstrated increased(18)F-FDG uptake consistent with an adenoma in 4 patients (40%; adenoma size range 3-14 mm). Maximum SUV was significantly higher for(18)F-FDG hrPET-positive tumors (difference = 5.1, 95% CI 2.1-8.1; p = 0.004) than for(18)F-FDG hrPET-negative tumors.(18)F-FDG hrPET positivity was not associated with tumor volume (p = 0.2) or dural invasion (p = 0.5). Midnight and morning ACTH levels were associated with(18)F-FDG hrPET positivity (p = 0.01 and 0.04, respectively) and correlated with the maximum SUV (R = 0.9; p = 0.001) and average SUV (R = 0.8; p = 0.01). All(18)F-FDG hrPET-positive adenomas had a less than a 180% ACTH increase and(18)F-FDG hrPET-negative adenomas had a greater than 180% ACTH increase after CRH stimulation (p = 0.03). Three adenomas were detected on SPGR MRI sequences that were not detected by(18)F-FDG hrPET imaging. Two adenomas not detected on SE (but no adenomas not detected on SPGR) were detected on(18)F-FDG hrPET. CONCLUSIONS While(18)F-FDG hrPET imaging can detect small functioning corticotroph adenomas and is more sensitive than SE MRI, SPGR MRI is more sensitive than(18)F-FDG hrPET and SE MRI in the detection of CD-associated pituitary adenomas. Response to CRH stimulation can predict(18)F-FDG hrPET-positive adenomas in CD.
| S-EPMC5091074 | biostudies-literature
Project description:The secreting function of pituitary adenomas (PAs) plays a critical role in making the treatment strategies. However, Magnetic Resonance Imaging (MRI) analysis for pituitary adenomas is labor intensive and highly variable among radiologists. In this work, by applying convolutional neural network (CNN), we built a segmentation and classification model to help distinguish functioning pituitary adenomas from non-functioning subtypes with 3D MRI images from 185 patients with PAs (two centers). Specifically, the classification model adopts the concept of transfer learning and uses the pre-trained segmentation model to extract deep features from conventional MRI images. As a result, both segmentation and classification models obtained high performance in two internal validation datasets and an external testing dataset (for segmentation model: Dice score = 0.8188, 0.8091 and 0.8093 respectively; for classification model: AUROC = 0.8063, 0.7881 and 0.8478, respectively). In addition, the classification model considers the attention mechanism for better model interpretation. Taken together, this work provides the first deep learning-based tumor region segmentation and classification models of PAs, which enables early diagnosis and subtyping PAs from MRI images.
| S-EPMC8178077 | biostudies-literature
Project description:There has been increasing experience in the utilization of intraoperative magnetic resonance imaging (iMRI) for intracranial surgery. Despite this trend, only a few U.S centers have examined the use of this technology for transsphenoidal resection of tumors of the sella. We present the largest series in North America examining the role of iMRI for pituitary adenoma resection. We retrospectively reviewed our institutional experience of 59-patients who underwent transsphenoidal procedures for sellar and suprasellar tumors with iMRI guidance. Of these, 52 patients had a histological diagnosis of pituitary adenoma. The technical results of this subgroup were examined. A 1.5-T iMRI was integrated with the BrainLAB (Feldkirchen, Germany) neuronavigation system. The majority (94%) of tumors in our series were macroadenomas. Seventeen percent of tumors were confined to the sella, 49% had suprasellar extensions without involvement of the cavernous sinus, 34% had frank cavernous sinus invasion. All patients underwent at least one iMRI, and 19% required one or more additional sets of intraoperative imaging. In 58% of patients, iMRI led to the surgeon attempting more resection. A gross total resection was obtained in 67% of the patients with planned total resections. There was one case of permanent postoperative diabetes insipidus and no other instances of new hormone replacement. In summary, iMRI was most useful for tumors of the sella with and without suprasellar extension where the information from the iMRI extended the complete resection rate from 40 to 72% and 55 to 88%, respectively. As one would expect, it did not substantially increase the rate of resection of tumors with cavernous sinus invasion. Overall, iMRI was particularly useful in guiding resection safely, aiding in clinical decision making, and allowing identification and preservation of the pituitary stalk and normal pituitary gland. Limitations of the iMRI include a need for additional personnel and training as well as additional operative time, which diminishes over time as personnel learn to optimize workflow efficiency. Additional costs are mitigated in part by using the iMRI as an immediate postoperative scan. Other data emerging from our experience suggest that preservation of normal gland and thus avoidance of hypopituitarism may be improved by iMRI use, but longer follow-up periods are required to test this conclusion. iMRI can detect unsuspected complications sooner than routine postoperative imaging, potentially leading to improved outcomes. However, larger studies are needed.
| S-EPMC3312115 | biostudies-other
Project description:Background:Endoscopic endonasal transsphenoidal surgery (EETS) is a standard technique used to approach sellar tumors. It is relatively safe, minimally invasive and carries a low risk of complications. However, one of the common complications reported with this technique is CSF leakage which causes morbidity, an increase in recovery time and hospital costs. This complication usually occurs from violation of the diaphragma sellae and a defect in the structures of the sellar floor or incomplete repair. In this article we report the first case with the use of a novel bilaminar chitosan scaffold which can be potentially used in the repair of the sellar floor, primarily aiming to the bony part of this structure. Case Presentation:After a personalized design employing a tissue engineering strategy, we reconstructed the sellar floor in a 65-year-old woman who had undergone EETS for a pituitary adenoma with progressive bilateral visual loss. To repair the bony defect of the sellar floor, we used a novel bilaminar chitosan scaffold. The patient had an unremarkable postoperative course with no evidence of CSF leak. The polymer was well tolerated without toxicity, infection or complications. After 2 years of follow up the patient remains neurologically intact, and in good endocrinological status. Conclusion:This is the first report of the use of this biomaterial and its biocompatibility in a clinical setting for the repair of the sellar floor during EETS. Our experience with chitosan bilaminar scaffold and in several preclinical studies in the literature have demonstrated good biocompatibility and effective bioengineered bone regeneration due to its excellent osteoconductive properties, this study pretends to be one landmark for further clinical research and larger case series with the use of this personalized tissue engineering materials in order to see they real efficacy to increase the surgeon armamentarium.
| S-EPMC7051988 | biostudies-literature
Project description:Machine learning and pattern recognition methods have been used to diagnose Alzheimer's disease (AD) and mild cognitive impairment (MCI) from individual MRI scans. Another application of such methods is to predict clinical scores from individual scans. Using relevance vector regression (RVR), we predicted individuals' performances on established tests from their MRI T1 weighted image in two independent data sets. From Mayo Clinic, 73 probable AD patients and 91 cognitively normal (CN) controls completed the Mini-Mental State Examination (MMSE), Dementia Rating Scale (DRS), and Auditory Verbal Learning Test (AVLT) within 3months of their scan. Baseline MRI's from the Alzheimer's disease Neuroimaging Initiative (ADNI) comprised the other data set; 113 AD, 351 MCI, and 122 CN subjects completed the MMSE and Alzheimer's Disease Assessment Scale-Cognitive subtest (ADAS-cog) and 39 AD, 92 MCI, and 32 CN ADNI subjects completed MMSE, ADAS-cog, and AVLT. Predicted and actual clinical scores were highly correlated for the MMSE, DRS, and ADAS-cog tests (P<0.0001). Training with one data set and testing with another demonstrated stability between data sets. DRS, MMSE, and ADAS-Cog correlated better than AVLT with whole brain grey matter changes associated with AD. This result underscores their utility for screening and tracking disease. RVR offers a novel way to measure interactions between structural changes and neuropsychological tests beyond that of univariate methods. In clinical practice, we envision using RVR to aid in diagnosis and predict clinical outcome.
| S-EPMC2871976 | biostudies-literature
Project description:Cardiovascular resonance (CMR) imaging is a standard imaging modality for assessing cardiovascular diseases (CVDs), the leading cause of death globally. CMR enables accurate quantification of the cardiac chamber volume, ejection fraction and myocardial mass, providing information for diagnosis and monitoring of CVDs. However, for years, clinicians have been relying on manual approaches for CMR image analysis, which is time consuming and prone to subjective errors. It is a major clinical challenge to automatically derive quantitative and clinically relevant information from CMR images.Deep neural networks have shown a great potential in image pattern recognition and segmentation for a variety of tasks. Here we demonstrate an automated analysis method for CMR images, which is based on a fully convolutional network (FCN). The network is trained and evaluated on a large-scale dataset from the UK Biobank, consisting of 4,875 subjects with 93,500 pixelwise annotated images. The performance of the method has been evaluated using a number of technical metrics, including the Dice metric, mean contour distance and Hausdorff distance, as well as clinically relevant measures, including left ventricle (LV) end-diastolic volume (LVEDV) and end-systolic volume (LVESV), LV mass (LVM); right ventricle (RV) end-diastolic volume (RVEDV) and end-systolic volume (RVESV).By combining FCN with a large-scale annotated dataset, the proposed automated method achieves a high performance in segmenting the LV and RV on short-axis CMR images and the left atrium (LA) and right atrium (RA) on long-axis CMR images. On a short-axis image test set of 600 subjects, it achieves an average Dice metric of 0.94 for the LV cavity, 0.88 for the LV myocardium and 0.90 for the RV cavity. The mean absolute difference between automated measurement and manual measurement is 6.1 mL for LVEDV, 5.3 mL for LVESV, 6.9 gram for LVM, 8.5 mL for RVEDV and 7.2 mL for RVESV. On long-axis image test sets, the average Dice metric is 0.93 for the LA cavity (2-chamber view), 0.95 for the LA cavity (4-chamber view) and 0.96 for the RA cavity (4-chamber view). The performance is comparable to human inter-observer variability.We show that an automated method achieves a performance on par with human experts in analysing CMR images and deriving clinically relevant measures.
| S-EPMC6138894 | biostudies-other
Project description:The Convolutional Neural Network algorithm achieved a sensitivity of 94% and specificity of 93% in identifying scans with vertebral fractures (VFs). The external validation results suggest that the algorithm provides an opportunity to aid radiologists with the early identification of VFs in routine CT scans of abdomen and chest.PurposeTo evaluate the performance of a previously trained Convolutional Neural Network (CNN) model to automatically detect vertebral fractures (VFs) in CT scans in an external validation cohort.MethodsTwo Chinese studies and clinical data were used to retrospectively select CT scans of the chest, abdomen and thoracolumbar spine in men and women aged ≥50 years. The CT scans were assessed using the semiquantitative (SQ) Genant classification for prevalent VFs in a process blinded to clinical information. The performance of the CNN model was evaluated against reference standard readings by the area under the receiver operating characteristics curve (AUROC), accuracy, Cohen's kappa, sensitivity, and specificity.ResultsA total of 4,810 subjects were included, with a median age of 62 years (IQR 56-67), of which 2,654 (55.2%) were females. The scans were acquired between January 2013 and January 2019 on 16 different CT scanners from three different manufacturers. 2,773 (57.7%) were abdominal CTs. A total of 628 scans (13.1%) had ≥1 VF (grade 2-3), representing 899 fractured vertebrae out of a total of 48,584 (1.9%) visualized vertebral bodies. The CNN's performance in identifying scans with ≥1 moderate or severe fractures achieved an AUROC of 0.94 (95% CI: 0.93-0.95), accuracy of 93% (95% CI: 93%-94%), kappa of 0.75 (95% CI: 0.72-0.77), a sensitivity of 94% (95% CI: 92-96%) and a specificity of 93% (95% CI: 93-94%).ConclusionThe algorithm demonstrated excellent performance in the identification of vertebral fractures in a cohort of chest and abdominal CT scans of Chinese patients ≥50 years.
| S-EPMC10786735 | biostudies-literature