The additional value of the lateral chest radiograph for the detection of small pulmonary nodules-a ROC analysis

The effect of C-reactive protein on chest X-ray interpretation: A decision-making experiment among pediatricians

INTRODUCTION

Clinical decision-making is complex and requires the integration of multiple sources of information. Physicians tend to over-rely on objective measures, despite the lack of supportive evidence in many cases. We sought to test if pediatricians over-rely on C-reactive protein (CRP) results when managing a child with suspected pneumonia.

METHODS

A nationwide decision-making experiment was conducted among 337 pediatricians in Israel. Each participant was presented with two detailed vignettes of a child with suspected pneumonia, each with a chest X-ray (CXR) taken from a real-life case of viral pneumonia. Participants were randomly assigned to one of three groups: Controls-where no lab tests were provided, and two intervention groups where the vignettes also noted a high or a low CRP value, in varying orders. Between-participant and within-participant analyses were conducted to study the effect of CRP on CXR interpretation. The three groups were presented with identical medical history, vital signs, findings on physical examination, blood count, and CXR.

RESULTS

Three-hundred and one pediatricians (89.3% of those approached) completed the study. Pediatricians were 60%-90% more likely to diagnose viral pneumonia as bacterial when presented with high CRP levels versus low CRP levels, despite the identical clinical data and CXR (62% vs. 39% and 58% vs. 31% of physicians; p = .002). Accordingly, they were 60%-90% more likely to prescribe antibiotics in these cases (86% vs. 53% and 78% vs. 41% of physicians; p < .001).

CONCLUSIONS

CRP by itself may modify the way in which pediatricians interpret a CXR, leading to the overprescription of antibiotics.

Reducing Annotation Burden Through Multimodal Learning

Choosing an optimal data fusion technique is essential when performing machine learning with multimodal data. In this study, we examined deep learning-based multimodal fusion techniques for the combined classification of radiological images and associated text reports. In our analysis, we (1) compared the classification performance of three prototypical multimodal fusion techniques: Early, Late, and Model fusion, (2) assessed the performance of multimodal compared to unimodal learning; and finally (3) investigated the amount of labeled data needed by multimodal vs. unimodal models to yield comparable classification performance. Our experiments demonstrate the potential of multimodal fusion methods to yield competitive results using less training data (labeled data) than their unimodal counterparts. This was more pronounced using the Early and less so using the Model and Late fusion approaches. With increasing amount of training data, unimodal models achieved comparable results to multimodal models. Overall, our results suggest the potential of multimodal learning to decrease the need for labeled training data resulting in a lower annotation burden for domain experts.

Oblique views of chest radiography from a designed rotation angle recommendation increase the contrast ratio between obscured lesions and surrounding structures

Background

Chest radiography (CXR) is the main tool used to detect pulmonary nodules. Lateral views of CXR are less effective and the aim of our study was to develop a rotation angle recommendation model to obtain the best oblique CXR with significantly increased contrast between lesions and surrounding normal structures in order to enhance the detection rate for potential obscured lesions on traditional posterior and anterior (PA) CXR.

Methods

A total of 140 subjects receiving low‐dose lung computed tomography (CT) screening were enrolled from the health check‐up database. An additional 14 cases with lung lesions on chest CT were included. Demography was also reviewed. Gross, left and right cardiothoracic ratios (CTR) were measured. All CT images were transformed to CXR to detect the best rotation angles and produce different views of CXR. Contrast ratio was calculated in the transformed CXR from CT with lesions. Comparison of contrast ratio among oblique, posterior‐anterior and lateral views was performed.

Results

CXR shows smaller gross CTR and left CTR but larger heart width and thoracic width in men than in women. Correlation evaluation displays gross CTR, heart width and left CTR are positively correlated with age only for the women group. The most important factor for the best rotation angle is right CTR for left rotation angle and left CTR for right rotation angle. The contrast ratio of the lesion to surrounding structures is significantly better on the oblique views in the designed angles than that on the traditional views.

Conclusion

CXR oblique views in the assigned angle from the 10‐degree rotation angle recommendation are able to enhance contrast ratio between the possible obscured lesions and surrounding structures on CXR.

The forgotten view: Chest X-ray - Lateral view

With CT (computed tomography) chest gaining more importance as a diagnostic tool, chest X-ray especially the lateral view is taken less commonly nowadays. Besides CT chest is also proven to be superior to chest X-ray in patients with major blunt trauma. We are presenting a 68-year old male who was partially treated from outside for a left sided pneumonia. He came to our hospital because of persisting chest pain. Chest X-ray, frontal view (postero-anterior) was almost normal except for a mild opacity in the left lower zone. CT scan of the chest revealed a fluid collection posteriorly enclosed within enhancing pleura. Chest X-ray, left lateral view showed a corresponding posterior pleural based opacity. We are presenting this case to highlight the importance of the lateral view of the chest X-ray. In selected cases there is still a role for the lateral view. With the three dimensional visualization provided by the CT, the lateral view of the chest may be easier to understand. Consequent to the initial diagnosis by CT further follow up can be done with the chest X-ray. In a limited way this mitigates unnecessary expenditure and more importantly prevents the patient from exposure to harmful radiation in the form of repeated CT.