Measurement of Endotracheal Tube Positioning on Chest X-Ray Using Object Detection.

Patients who are intubated with endotracheal tubes often receive chest x-ray (CXR) imaging to determine whether the tube is correctly positioned. When these CXRs are interpreted by a radiologist, they evaluate whether the tube needs to be repositioned and typically provide a measurement in centimeters between the endotracheal tube tip and carina. In this project, a large dataset of endotracheal tube and carina bounding boxes was annotated on CXRs, and a machine-learning model was trained to generate these boxes on new CXRs and to calculate a distance measurement between the tube and carina. This model was applied to a gold standard annotated dataset, as well as to all prospective data passing through our radiology system for two weeks. Inter-radiologist variability was also measured on a test dataset. The distance measurements for both the gold standard dataset (mean error = 0.70 cm) and prospective dataset (mean error = 0.68 cm) were noninferior to inter-radiologist variability (mean error = 0.70 cm) within an equivalence bound of 0.1 cm. This suggests that this model performs at an accuracy similar to human measurements, and these distance calculations can be used for clinical report auto-population and/or worklist prioritization of severely malpositioned tubes.

Abdominal and pelvic complications of nonoperative oncologic therapy.

Oncologic patients are treated with a combination of chemotherapy, radiation therapy, and surgery. Advances in therapeutic options have greatly improved the survival of patients with cancer. Examples of these advances are newer chemotherapeutic agents that target the cell receptors and advanced radiation therapy delivery systems. It is imperative that radiologists be aware of the variety of imaging findings seen after therapy in patients with cancer. Complications may occur with classic cytotoxic therapies (eg, 5-fluorouracil), usually at higher or prolonged doses or when administered to radiosensitive areas. Newer targeted systemic agents, such as bevacizumab and imatinib, have associated characteristic toxicities because their effects on cells do not depend on dose. Radiation may induce early and late effects in local normal tissues that may be seen at imaging. Imaging findings after chemotherapy include fatty liver, pseudocirrhosis, hepatic veno-occlusive disease, and splenic rupture. Complications of radiation therapy include large and small bowel strictures and radiation-induced hepatitis and tumors. Awareness of the various therapeutic options and knowledge of the spectrum of posttherapeutic complications allows radiologists to provide a comprehensive report that may impact patient management.

A multi-institutional meningioma MRI dataset for automated multi-sequence image segmentation.

Meningiomas are the most common primary intracranial tumors and can be associated with significant morbidity and mortality. Radiologists, neurosurgeons, neuro-oncologists, and radiation oncologists rely on brain MRI for diagnosis, treatment planning, and longitudinal treatment monitoring. However, automated, objective, and quantitative tools for non-invasive assessment of meningiomas on multi-sequence MR images are not available. Here we present the BraTS Pre-operative Meningioma Dataset, as the largest multi-institutional expert annotated multilabel meningioma multi-sequence MR image dataset to date. This dataset includes 1,141 multi-sequence MR images from six sites, each with four structural MRI sequences (T2-, T2/FLAIR-, pre-contrast T1-, and post-contrast T1-weighted) accompanied by expert manually refined segmentations of three distinct meningioma sub-compartments: enhancing tumor, non-enhancing tumor, and surrounding non-enhancing T2/FLAIR hyperintensity. Basic demographic data are provided including age at time of initial imaging, sex, and CNS WHO grade. The goal of releasing this dataset is to facilitate the development of automated computational methods for meningioma segmentation and expedite their incorporation into clinical practice, ultimately targeting improvement in the care of meningioma patients.

Detection of Critical Spinal Epidural Lesions on CT Using Machine Learning.

BACKGROUND: Critical spinal epidural pathologies can cause paralysis or death if untreated. Although magnetic resonance imaging is the preferred modality for visualizing these pathologies, computed tomography (CT) occurs far more commonly than magnetic resonance imaging in the clinical setting. OBJECTIVE: A machine learning model was developed to screen for critical epidural lesions on CT images at a large-scale teleradiology practice. This model has utility for both worklist prioritization of emergent studies and identifying missed findings. MATERIALS AND METHODS: There were 153 studies with epidural lesions available for training. These lesions were segmented and used to train a machine learning model. A test data set was also created using previously missed epidural lesions. The trained model was then integrated into a teleradiology workflow for 90 days. Studies were sent to secondary manual review if the model detected an epidural lesion but none was mentioned in the clinical report. RESULTS: The model correctly identified 50.0% of epidural lesions in the test data set with 99.0% specificity. For prospective data, the model correctly prioritized 66.7% of the 18 epidural lesions diagnosed on the initial read with 98.9% specificity. There were 2.0 studies flagged for potential missed findings per day, and 17 missed epidural lesions were found during a 90-day time period. These results suggest almost half of critical spinal epidural lesions visible on CT imaging are being missed on initial diagnosis. CONCLUSION: A machine learning model for identifying spinal epidural hematomas and abscesses on CT can be implemented in a clinical workflow.

PET/CT in the management of patients with stage IIIC and IV metastatic melanoma considered candidates for surgery: evaluation of the additive value after conventional imaging.

OBJECTIVE: The purpose of this article is to determine how often unexpected (18)F-FDG PET/CT findings result in a change in management for patients with stage IV and clinically evident stage III melanoma with resectable disease according to conventional imaging. SUBJECTS AND METHODS: Thirty-two patients with oligometastatic stage IV and clinically evident stage III melanoma were identified by surgical oncologists according to the results of conventional imaging, which included contrast-enhanced CT of the chest, abdomen, and pelvis and MRI of the brain. The surgical plan included resection of known metastases or isolated limb perfusion with chemotherapy. Thirty-three FDG PET/CT scans were performed within 36 days of their contrast-enhanced CT. The impact of PET/CT was defined as the percentage of cases in which a change in the surgical plan resulted from the unanticipated PET/CT findings. RESULTS: PET/CT revealed unexpected melanoma metastases in 12% of scans (4/33). As a result, the surgery was canceled for two patients, and the planned approach was altered for another two patients to address the unexpected sites. In 6% of scans (2/33), the unexpected metastases were detected in the extremities, which were not included in conventional imaging. Three scans (9%) showed false-positive FDG-avid findings that proved to be benign by subsequent stability or resolution with no therapy. CONCLUSION: In patients with surgically treatable metastatic melanoma, FDG PET/CT can detect unexpected metastases that are missed or not imaged with conventional imaging, and can be considered as part of preoperative workup.

Radiologic manifestations of immune-related adverse events in patients with metastatic melanoma undergoing anti-CTLA-4 antibody therapy.

OBJECTIVE: Monoclonal antibodies against cytotoxic T-lymphocyte antigen 4 (CTLA-4) used for treatment of metastatic melanoma produce inflammatory immune-related adverse events. The purpose of the current study was to retrospectively identify and characterize the radiologic manifestations of immune-related adverse events and to evaluate the possible association between these events and clinical responses to anti-CTLA-4 therapy. MATERIALS AND METHODS: We retrospectively reviewed the images and medical records of 119 patients with metastatic melanoma treated with anti-CTLA-4 at our institution and assessed the presence of radiologic manifestations of immune-related adverse events and the clinical responses to therapy. The responses were categorized as progressive or controlled disease. The controlled disease category included stable disease, partial response, and complete response according to the Response Evaluation Criteria in Solid Tumors, version 1.1. RESULTS: Radiologic manifestations of immune-related adverse events were found in 20 patients (16.8%). Clinically evident manifestations included colitis, hypophysitis, thyroiditis, and arthritis. Clinically silent manifestations were benign lymphadenopathy and inflammatory changes in the soft tissues, such as myositis, fasciitis, and retroperitoneal fat haziness. There was a significant association between the incidence of radiologic manifestations of immune-related adverse events and clinical responses to anti-CTLA-4 therapy. The disease control rates were 18% for the entire group, 55% for the group with, and 10% for the group without radiologic manifestations of immune-related adverse events. In three patients (2.5%), lymphadenopathy related to radiologic manifestations of immune-related adverse events was interpreted as suspected metastasis but was proved benign at biopsy. CONCLUSION: Radiologic manifestations of immune-related adverse events are associated with significant clinical benefit of anti-CTLA-4 therapy. In the era of developing immune checkpoint-targeted therapy for metastatic melanoma, radiologists should be alert to the possibility of these manifestations, which can mimic radiologic disease progression.

Evaluation of cancer treatment in the abdomen: Trends and advances.

Response evaluation in Oncology has relied primarily on change in tumor size. Inconsistent results in the prediction of clinical outcome when size based criteria are used and the increasing role of targeted and loco-regional therapies have led to the development of new methods of response evaluation that are unrelated to change in tumor size. The goals of this review are to expose briefly the size based criteria and to present the non-size based approaches that are currently applicable in the clinical setting. Other paths that are still being explored are not discussed in details.

F-18 FDG PET/CT for detection of malignant involvement of peripheral nerves: case series and literature review.

PURPOSE: To evaluate the role of positron emission tomography plus computed tomography (PET/CT) scans in detecting malignant involvement of the peripheral nerves (PNs). MATERIAL AND METHODS: We retrospectively reviewed medical records of all PET/CT studies performed at The University of Texas MD Anderson Cancer Center between 2003 and 2009, and selected patients in whom F-18 fluorodeoxyglucose (FDG) PET/CT findings were suspicious for malignant involvement of the PNs. We identified 26 cases of suspected tumorous involvement of the PNs that was subsequently confirmed by either biopsy or clinical follow-up. We evaluated the value of PET/CT in diagnosing malignant involvement of the PNs. RESULTS: Of the 26 patients, 12 had lymphoma, 10 had breast cancer, 2 had lung cancer, 1 had colon cancer, and 1 had melanoma. In 21 patients, magnetic resonance imaging (MRI) was performed, either for follow-up of the PET/CT finding or to find an explanation for symptoms. MRI confirmed the presence of disease in only 9 patients, was interpreted as normal in 7 patients, and was inconclusive in 5 patients. FDG PET/CT was able to differentiate an active tumor from post-treatment fibrosis and could assess response to therapy with a high degree of confidence. CONCLUSIONS: Our results indicate that FDG PET/CT is helpful in diagnosing malignant involvement of the PNs, especially when findings from anatomic imaging (MRI or CT) are negative. In cases of known treated malignancy involving the PNs, follow-up by PET/CT has the advantage of high sensitivity for local recurrence.

Contemporary diagnostic imaging modalities for the staging and surveillance of melanoma patients: a meta-analysis.

BACKGROUND: Meta-analyses were performed to examine the utility of ultrasonography, computed tomography (CT), positron emission tomography (PET), and a combination of both (PET-CT) for the staging and surveillance of melanoma patients. METHOD: Patient-level data from 74 studies containing 10,528 patients (between January 1, 1990, and June, 30, 2009) were used to derive characteristics of the diagnostic tests used. Meta-analyses were conducted by use of Bayesian bivariate binomial models to estimate sensitivity and specificity. Diagnostic odds ratios [ie, true-positive results/false-negative results)/(false-positive results/true-negative results)] and their 95% credible intervals (CrIs) and positive predictive values were used as indicators of test performance. RESULTS: Among the four imaging methods examined for the staging of regional lymph nodes, ultrasonography had the highest sensitivity (60%, 95% CrI = 33% to 83%), specificity (97%, 95% CrI = 88% to 99%), and diagnostic odds ratio (42, 95% CrI = 8.08 to 249.8). For staging of distant metastases, PET-CT had the highest sensitivity (80%, 95% CrI = 53% to 93%), specificity (87%, 95% CrI = 54% to 97%), and diagnostic odds ratio (25, 95% CrI = 3.58 to 198.7). Similar trends were observed for melanoma surveillance of lymph node involvement, with ultrasonography having the highest sensitivity (96%, 95% CrI = 85% to 99%), specificity (99%, 95% CrI = 95% to 100%), and diagnostic odds ratio (1675, 95% CrI = 226.6 to 15,920). For distant metastases, PET-CT had the highest sensitivity (86%, 95% CrI = 76% to 93%), specificity (91%, 95% CrI = 79% to 97%), and diagnostic odds ratio (67, 95% CrI = 20.42 to 229.7). Positive predictive values were likewise highest for ultrasonography in lymph node staging and for PET-CT in detecting distant metastases. CONCLUSION: Among the compared modalities, ultrasonography was superior for detecting lymph node metastases, and PET-CT was superior for the detection of distant metastases in both the staging and surveillance of melanoma patients.

Pelvic computed tomography scans for surveillance in patients with primary melanoma in the head and neck.

Although pelvic computed tomography (CT) scans are frequently performed as a part of routine surveillance, the evidence for or against the routine use of these scans in patients with primary melanoma in the head and neck is weak. We conducted a retrospective study to evaluate the value of pelvic CT scans as routine surveillance in patients with primary melanoma in the head and neck. We identified 146 patients with either primary or mucosal primary melanoma who had adequate follow-up evaluation for at least 5 years at our institution. Among them, 33 patients (23%) had stage III melanoma, and four (3%) had stage IV melanoma at the time of diagnosis. At a median follow-up duration of 49 months, 110 patients (75%) had developed recurrences, and the median time to the first recurrence was 13 months. A total of 82 (56%) patients had eventually developed distant metastases, but only 10 (7%) had developed metastases in the pelvis, and none had developed pelvic metastases as the first and the only site of recurrence. If the true rate of finding the pelvic metastasis as the first and the only recurrence was at least 3%, the probability of seeing 0 events of the 146 patients was 1.17%. This study, which is the largest series to evaluate the value of pelvic CT scans in this patient population to date, suggests that the routine use of a pelvic CT scan as a surveillance method does not have any impact on the management in patients with primary melanoma in the head and neck.

A phase II study of gefitinib in patients with metastatic melanoma.

Gefitinib is an inhibitor of the epidermal growth factor receptor, which is frequently expressed on both choroidal and nonchoroidal melanoma cells. We evaluated the clinical efficacy of gefitinib in patients with metastatic melanoma. Patients with stage IV or unresectable stage III melanoma and Zubrod performance status of less than or equal to 2 were eligible. Previous systemic treatment for metastatic disease was required. The dose of oral gefitinib was 250 mg administered daily, and tumor response was evaluated every 6 weeks. Forty-six patients with nonchoroidal melanoma and six with choroidal melanoma were treated, and 48 were evaluable for response. The median age was 62.5 years. Forty-one patients (79%) had stage M1c disease. There were no drug-related grade 4 or 5 adverse events, and fatigue was the only grade 3 adverse event that occurred in more than 5% of patients. Two patients (4%) had partial responses and 13 patients (27%) had disease stabilization. The two responders had a median duration of response of 10.9 months. The median overall progression-free survival was 1.4 months and the median overall survival was 9.7 months. Among the patients with sufficient tissues obtained before and 6 weeks after starting gefitinib administration, there were no notable trends in the changes of the tumoral expression of p-ERK1/2, p-AKT, PAK1, and serum levels of vascular endothelial growth factor or IL-8 with treatment. We concluded that gefitinib was well tolerated but had minimal clinical efficacy as a single-agent therapy for unselected patients with metastatic melanoma.

Treatment with tyrosine kinase inhibitors for patients with differentiated thyroid cancer: the M. D. Anderson experience.

OBJECTIVES: Until recently, treatment options for patients with progressive, radioactive iodine-resistant differentiated thyroid cancer (DTC) have been limited. In our clinical practice, we have begun to use sorafenib and sunitinib for patients with progressive DTC who are not able or willing to participate in clinical trials. In this paper, we describe the University of Texas M. D. Anderson Cancer Center's experience with the off-label use of these tyrosine kinase inhibitors for DTC. METHODS: Adult patients were included if they had a diagnosis of radioactive iodine-refractory DTC, were treated with single agent sorafenib or sunitinib, and had both baseline and at least one follow-up scan for restaging purposes. All imaging data were collected, as well as the TSH-suppressed thyroglobulin (Tg) levels corresponding to each scan date. The primary endpoints were radiographic response and progression-free survival (PFS). Secondary objectives were tissue-specific radiographic responses and correlation of Tg with overall response. RESULTS: We identified 33 patients from our clinical database. Fifteen patients (nine women, six men) met inclusion criteria, with a median age of 61 yr (range, 38-83 yr). Eight patients had papillary and seven had follicular thyroid carcinoma. Sorafenib was used in 13 and sunitinib in two, including one patient who failed prior sorafenib therapy. All patients had evidence of progressive disease (PD) before start of therapy, with a median PFS of only 4 months. Best response in target lesions was: partial response (PR) in three (20%), stable disease (SD) in nine (60%), and PD in three (20%). Clinical benefit (PR+SD) was 80%. The sunitinib patient previously refractory to sorafenib had a 38% reduction in tumor size. The most noticeable organ-specific response was observed in lung (median change, -22%) compared to lymph nodes (median change, 0%). Pleural disease and nonirradiated bone metastases demonstrated PD. All histological subtypes had similar responses. The median PFS was 19 months. The median overall survival has not yet been reached, but at 2 yr of follow-up, overall survival is 67%. Log Tg correlated with radiographic response (P = 0.0005). CONCLUSIONS: Sorafenib and sunitinib appear to be effective in patients with widely metastatic, progressive DTC, with most patients achieving SD or PR, despite having PD at baseline. The most noticeable responses occurred in the lungs in contrast with minimal changes in nodal metastases and PD in pleural and nonirradiated bone metastases, suggesting a tissue-specific response to therapy. Log Tg significantly correlated with response to treatment and therefore may have value as a surrogate marker of response.

Detection of small pancreatic tumors with multiphasic helical CT.

OBJECTIVE: The purpose of this study was to evaluate the sensitivity and specificity of helical CT in the detection of adenocarcinomas of the pancreas measuring 2 cm or smaller at pathologic examination. MATERIALS AND METHODS: Thin-section triple phase (20, 40, and 70 sec after the start of injection) contrast-enhanced helical CT scans of the abdomen in 18 patients with a pancreatic carcinoma that was 2 cm or smaller and 18 patients with a normal pancreas were retrospectively reviewed by two senior radiologists who specialized in oncologic abdominal imaging. Discrepancies were resolved by consensus. The observers were unaware of the clinical information. CT scans were evaluated for the presence of a pancreatic mass, bile, and pancreatic duct stricture. The location and size of tumors as determined on CT were compared with pathologic findings. The CT results were also compared with the prospective CT interpretations derived from the radiology reports and with the endoscopic sonographic reports when available. RESULTS: The sensitivity of thin-section triple-phase helical CT in the detection of small pancreatic masses was 77%, and the specificity was 100% for the two experienced observers. The sensitivity and specificity were 72% and 100%, respectively, for the prospective interpretations done by 10 observers. There was no correlation between the tumor size at pathology and the CT measurements. CONCLUSION: Thin-section contrast-enhanced helical CT is sensitive and highly specific for the detection of pancreatic tumors measuring 2 cm or smaller. Improvement in the detection rate of this technique compared with previous techniques lies in the optimization of parenchymal enhancement during the pancreatic phase and the decrease in slice thickness.