• AI
  • Molecular Imaging
  • CT
  • X-Ray
  • Ultrasound
  • MRI
  • Facility Management
  • Mammography

Clinical Roundup: 5 Studies You Can’t Miss This Month


From diagnosing renal cancer to the radiologist’s role in preventing intimate partner violence, here are the studies you shouldn’t miss this month.

Stack of articles

Contrast-Enhanced Ultrasound Offers Enhanced Diagnostic Sensitivity for Diagnosis of Renal Cancer

A new meta-analysis published in the Journal of Ultrasound Medicine suggests that contrast-enhanced ultrasound (CEUS) may offer more sensitivity and specificity for the diagnosis of renal cancer than contrast-enhanced computer tomography (CECT).

Researchers from China compared the diagnostic accuracy of CEUS to CECT in order to evaluate patients for renal cancer. The researchers reviewed data sets from twenty-two studies published before 2017, performing a meta-analysis on a variety of diagnostic factors in order to compare and contrast the diagnostic value of these two methods. Those factors included sensitivity and specificity, as well as:

  • Positive and negative likelihood ratios;

  • Positive and negative predictive values;

  • Diagnostic odds ratio;

  • Summary receiver operating characteristic curve and area under the cure;

  • and potential other sources of heterogeneity.

The results of the analysis showed CEUS to have strong diagnostic strength, with a summary sensitivity and specificity measured at 0.96 and 0.82, respectively, and a summary diagnostic odds ratio at 102.04. When the two methods were compared directly, CEUS demonstrated a higher diagnostic sensitivity (0.94) over CECT (0.85) when diagnosing renal cancer, when the specificities were comparable, leading the researchers to conclude that CEUS may be a better tool for diagnosing this form of cancer in patients.

Radiology has a Role to Play in Identifying Intimate Partner Violence

Intimate partner violence (IPV), sometimes referred to as domestic violence, is a serious, yet preventable public health concern. Yet, because victims often go to great lengths to hide their abuse from medical personnel, it can be difficult for physicians or other clinical staff to identify the problem. Now, a new study in the journal Radiology suggests that radiologic findings can help better identify victims before they leave the care setting, getting them the help they need.

Researchers from Brigham and Women’s Hospital and the Harvard T.H. Chan School of Public Health compared the electronic medical records from 185 patients who had been referred to an IPV support program after visiting the emergency room with more than 500 age- and sex-matched control records from patients who had also received some type of emergency care. They discovered that variables including race, previously diagnosed psychiatric conditions, and homelessness were highly correlated with IPV program referrals. But, in addition, these patients underwent more imaging studies for obstetric-gynecologic issues or acute fractures, particularly in the face or skull, than the matched control population.

In a related editorial article, radiologists from Massachusetts General stated that radiologists are often in a position to be the first to identify children who may be experiencing physical abuse in the home. By better understanding both the clinical and imaging features of IPV, as well as noting the number of imaging studies in the record for a particular patient, radiologists may be in a similarly unique position to recognize what is, too often, a fatal issue, and refer patients to IPV intervention programs before it is too late.

The Difference Between Rheumatoid Lung Nodules and Malignancies

Researchers from the Mayo Clinic have discovered distinct clinical and imaging differences in rheumatoid pulmonary nodules when compared to a malignant tumor, according to a new study published in the European Journal of Radiology.

A small subset of patients with rheumatoid arthritis-less than 1% of the population-may develop pulmonary nodules in the lung. Because these nodules can be quite variable in appearance, they are sometimes difficult to differentiate from a potential malignancy. In order to better distinguish the two different types of nodules, the researchers conducted a retrospective review of the CT and PET/CT scans for 73 patients, with mean age of 67 years, who were diagnosed with either rheumatoid or malignant lung nodules at the Mayo Clinic.

The researchers discovered that patients with rheumatoid lung nodules tend, on average, to be younger than those with malignancies, and also show subcutaneous rheumatoid nodules and seropositivity. In addition, they noted that certain distinct CT features were more commonly associated with rheumatoid lung nodules including:

  • Multiplicity;

  • Smooth border;

  • Cavitation;

  • Satellite nodules;

  • Pleural contact;

  • and a subpleural rind of soft tissue.

The group also discovered a similar group of distinct features when the nodules are scanned using FDG-PET/CT. These features include a low level of metabolism and a lack of draining lymph nodes to help remove the FDG agent.

Given these findings, the researchers concluded that radiologists can look to these features to help make more accurate diagnoses in patients who present with nodules in order to make sure they receive the appropriate treatment regimen.

AI Can Differentiate Benign and Malignant Breast Tumors

A new study published in the Japanese Journal of Radiology suggests a deep learning artificial intelligence (AI) algorithm can discriminate between ultrasound images of benign and malignant breast masses as well as-and, at times, perhaps even better than-trained radiologists.

Researchers from the Tokyo Medical and Dental University developed a convolutional neural network (CNN) using the GoogLeNet architecture, training the algorithm on nearly 500 ultrasound images of both benign and malignant breast masses. The researchers then tested the algorithm on a novel test data set which included images of 48 benign masses and 72 malignancies. In addition to running the test data through the AI network, the researchers also recruited three trained radiologists to review and interpret the images. They then measured the sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve (AUC) for both the algorithm and the physicians.

When compared to the radiologists, the group found that the AI program performed as well as, or even better than the doctors on diagnosing the two types of breast masses. They noted that the algorithm demonstrated:

  • A sensitivity of 0.958, as compared to the radiologists at 0.583-0.917;

  • A specificity of 0.925, as compared to the radiologists at 0.604-0.771;

  • and an accuracy of 0.925, as compared to the radiologists at 0.658-0.792.

The researchers concluded that this deep learning network showed high diagnostic performance-and could be used to augment radiologists’ capabilities in order to better detect breast mass malignancies in patients in the future.

Diagnosing Appendicitis: CT vs. MRI

When patients present with severe abdominal pain, emergency physicians generally rely on CT scans to diagnose-or rule out-appendicitis. In a new study in the Journal of Magnetic Resonance Imaging, researchers from BerbeeWalsh Department of Emergency Medicine and the University of Wisconsin-Madison, compared CT to MRI to see if MRI has the required sensitivity to be used in CT’s stead when dealing with appendicitis or other abdominopelvic pathologies.

The researchers used CT and MR images from 113 patients over the age of 12 years suspected of having appendicitis after coming to the emergency room and complaining of abdominal pain. The group recruited three radiologists to separately review and interpret the images, using a standard case report form to outline their diagnoses. The researchers then convened an expert panel to do a chart review of those case report forms in order to determine final diagnoses.

When the group measured sensitivity, as well as the sensitivity for consensus reads, for noncontrast-enhanced MRI, contrast-enhanced MRI, and CT for acute diagnoses, they found the following:

  • Noncontrast-enhanced MRI at 77% and 82%, respectively;

  • Contrast-enhanced MRI at 84.2% and 87.1%, respectively;

  • and CT at 88.7% and 92.2%, respectively.

The researchers concluded that contrast-enhanced MRI demonstrated similar sensitivity to CT when diagnosing abdominopelvic pathology, including appendicitis-suggesting that emergency departments can rely on this technique over CT, reducing exposure to unnecessary ionizing radiation to patients.

Related Videos
Where the USPSTF Breast Cancer Screening Recommendations Fall Short: An Interview with Stacy Smith-Foley, MD
A Closer Look at MRI-Guided Transurethral Ultrasound Ablation for Intermediate Risk Prostate Cancer
Improving the Quality of Breast MRI Acquisition and Processing
Does Initial CCTA Provide the Best Assessment of Stable Chest Pain?
Making the Case for Intravascular Ultrasound Use in Peripheral Vascular Interventions
Can Diffusion Microstructural Imaging Provide Insights into Long Covid Beyond Conventional MRI?
Emerging MRI and PET Research Reveals Link Between Visceral Abdominal Fat and Early Signs of Alzheimer’s Disease
Nina Kottler, MD, MS
Practical Insights on CT and MRI Neuroimaging and Reporting for Stroke Patients
Related Content
© 2024 MJH Life Sciences

All rights reserved.