Accuracy of Optic Nerve Sheath Diameter Measurements in Pocket-Sized Ultrasound Devices in a Simulation Model

The concurrent validity of a portable ultrasound probe for muscle thickness measurements

Ultrasound imaging is extensively used by both practitioners and researchers in assessing muscle thickness (MT); however, its use in the field is constrained by the transportability of stationary devices. New portable ultrasound probes pose as a cost-effective and transportable alternative for field-based assessments. This study evaluated the concurrent validity of a portable probe (Lumify) against a laboratory-based device (Vivid S5) in measuring MT. Eighteen participants (nine males and nine females) visited the laboratory and their MT measurements were collected using each device at five different sites (anterior and posterior arm, anterior and posterior thigh, and posterior lower leg). Bland-Altman plots (systematic and proportional bias, random error, and 95% limits of agreement), Pearson's product-moment correlation coefficient (r), and paired samples t-tests with Cohen's d effect sizes (ES) were used to assess the concurrent validity of the Lumify device. Systematic bias was low at all sites ( ≤ 0.11 cm) while proportional bias was detected only at the posterior lower leg (r2 = 0.217 [r = 0.466]). The difference in MT between devices was significant only at the anterior thigh (p < 0.05); however, ES for all sites were considered trivial (ES ≤ 0.131). Linear associations were found between the devices at each site of measurement (r ≥ 0.95). These results highlight that the Lumify probe can be used interchangeably with the Vivid S5 for MT measurements, providing practitioners and researchers with a more cost-effective and portable alternative for field-based assessments.

Development of a Deep Learning-Based System for Optic Nerve Characterization in Transorbital Ultrasound Images on a Multicenter Data Set

OBJECTIVE

Characterization of the optic nerve through measurement of optic nerve diameter (OND) and optic nerve sheath diameter (ONSD) using transorbital sonography (TOS) has proven to be a useful tool for the evaluation of intracranial pressure (ICP) and multiple neurological conditions. We describe a deep learning-based system for automatic characterization of the optic nerve from B-mode TOS images by automatic measurement of the OND and ONSD. In addition, we determine how the signal-to-noise ratio in two different areas of the image influences system performance.

METHODS

A UNet was trained as the segmentation model. The training was performed on a multidevice, multicenter data set of 464 TOS images from 110 subjects. Fivefold cross-validation was performed, and the training process was repeated eight times. The final prediction was made as an ensemble of the predictions of the eight single models. Automatic OND and ONSD measurements were compared with the manual measurements taken by an expert with a graphical user interface that mimics a clinical setting.

RESULTS

A Dice score of 0.719 ± 0.139 was obtained on the whole data set merging the test folds. Pearson's correlation was 0.69 for both OND and ONSD parameters. The signal-to-noise ratio was found to influence segmentation performance, but no clear correlation with diameter measurement performance was determined.

CONCLUSION

The developed system has a good correlation with manual measurements, proving that it is feasible to create a model capable of automatically analyzing TOS images from multiple devices. The promising results encourage further definition of a standard protocol for the automatization of the OND and ONSD measurement process using deep learning-based methods. The image data and the manual measurements used in this work will be available at 10.17632/kw8gvp8m8x.1.

Non-Invasive Intracranial Pressure Monitoring

(1) Background: Intracranial pressure (ICP) monitoring plays a key role in the treatment of patients in intensive care units, as well as during long-term surgeries and interventions. The gold standard is invasive measurement and monitoring via ventricular drainage or a parenchymal probe. In recent decades, numerous methods for non-invasive measurement have been evaluated but none have become established in routine clinical practice. The aim of this study was to reflect on the current state of research and shed light on relevant techniques for future clinical application. (2) Methods: We performed a PubMed search for “non-invasive AND ICP AND (measurement OR monitoring)” and identified 306 results. On the basis of these search results, we conducted an in-depth source analysis to identify additional methods. Studies were analyzed for design, patient type (e.g., infants, adults, and shunt patients), statistical evaluation (correlation, accuracy, and reliability), number of included measurements, and statistical assessment of accuracy and reliability. (3) Results: MRI-ICP and two-depth Doppler showed the most potential (and were the most complex methods). Tympanic membrane temperature, diffuse correlation spectroscopy, natural resonance frequency, and retinal vein approaches were also promising. (4) Conclusions: To date, no convincing evidence supports the use of a particular method for non-invasive intracranial pressure measurement. However, many new approaches are under development.

Agreement and reliability of lower limb muscle architecture measurements using a portable ultrasound device

High end ultrasonography devices lack in portability and are expensive. We investigated the agreement and reliability of a handheld and portable ultrasound system for human lower limb muscle architecture measurements. We captured ultrasound images of the rectus femoris (RF), vastus lateralis (VL) and gastrocnemius medialis (GM) in 36 active healthy participants (15 female, 21 male) at 50% of muscle length using the handheld Lumify (L12-4, linear-array 37 mm, Philips Healthcare, Amsterdam, Netherlands) and a high-end laboratory device (ACUSON Juniper, linear-array 54 mm, 12L3, SIEMENS Healthineers, Erlangen, Germany). We compared measurements of muscle fascicle length, pennation angle and thickness. To assess inter-session reliability of the Lumify system, participants were measured twice within 1 week. Comparing RF architecture measurements of both devices resulted in intra-class correlations (ICCs) ranging from 0.46–0.82 and standardized mean difference (SMDs) ranging from −0.45–0.05. For VL, ICCs ranged from 0.60–0.89 and SMDs ranged from −0.11–0.13. ICCs and SMDs for the GM ranged from 0.82–0.86 and −0.07–0.07. Calculating inter-session reliability for RF resulted in ICCs ranging from 0.44–0.76 and SMDs ranging from −0.38–0.15. For VL, ICCs and SMDs ranged from 0.57–0.75 and −0.13–0.02. ICCs for GM ranged from 0.75–0.92 and SMDs ranged from −0.15–0.16. Measurement of muscle thickness demonstrated the highest agreement (ICC ≥0.82) and reliability (ICC ≥0.75) across all muscles. The Lumify system was comparable to a high-end device and reliable for GM measurements. However, agreement and reliability were lower for the RF and VL. Of all evaluated architectural parameters, muscle thickness exhibited highest agreement and reliability.