Automatic segmentation of ultrasound images of gastrocnemius medialis with different echogenicity levels using convolutional neural networks.

The purpose of this study was to develop an automatic method for the segmentation of muscle cross-sectional area on transverse B-mode ultrasound images of gastrocnemius medialis using a convolutional neural network(CNN). In the provided dataset images with both normal and increased echogenicity are present. The manually annotated dataset consisted of 591 images, from 200 subjects, 400 relative to subjects with normal echogenicity and 191 to subjects with augmented echogenicity. From the DICOM files, the image has been extracted and processed using the CNN, then the output has been post-processed to obtain a finer segmentation. Final results have been compared to the manual segmentations. Precision and Recall scores as mean ± standard deviation for training, validation, and test sets are 0.96 ± 0.05, 0.90 ± 0.18, 0.89 ± 0.15 and 0.97 ±0.03, 0.89± 0.17, 0.90 ± 0.14 respectively. The CNN approach has also been compared to another automatic algorithm, showing better performances. The proposed automatic method provides an accurate estimation of muscle cross-sectional area in muscles with different echogenicity levels.

Simple surface EMG recording as a noninvasive screening method for the detection of acute oxaliplatin-induced neurotoxicity: a feasibility pilot study.

OBJECTIVES: Oxaliplatin-induced neurotoxicity can be a dose-limiting side effect to effective chemotherapy. Acute hyperexcitability causes cold-evoked sensory and motor symptoms, which resemble neuromyotonia. An accessible and non-invasive technique for early detection could help select patients for potential treatments. We assessed the use of a simple surface electromyography (sEMG) in patients directly after oxaliplatin infusion. METHODS: In patients with colorectal cancer, acute neurotoxicity was evaluated by means of a physical examination, a questionnaire, and sEMG directly after the second and fourth cycle of oxaliplatin. Questionnaires were also assessed 1 day after infusion. RESULTS: 14 patients were measured after the second cycle and 8 patients were also measured after the fourth cycle of oxaliplatin. All patients reported to a variable degree oxaliplatin induced neurotoxicity symptoms: sensitivity to touching cold or swallowing cold items were reported as most severe. Clinical signs of hyperexcitability were observed in 55% of the measurements. Spontaneous activity compatible with neuromyotonia was observed in 82% of the sEMG recordings. CONCLUSIONS: Patient reported symptoms, physical examination and simple sEMG are complementary measurements to detect acute oxaliplatin induced neurotoxicity. After further validation, sEMG recording can be used as a simple objective screenings tool to detect nerve hyperexcitability directly after oxaliplatin administration.

Quantitative facial muscle ultrasound: feasibility and reproducibility.

INTRODUCTION: In this study we describe a protocol for quantitative ultrasound of facial muscles (procerus, zygomaticus major, levator labii superior, depressor anguli oris, mentalis, orbicularis oris pars labialis, orbicularis oris pars marginalis). METHODS: Muscle thickness (MT) and echo intensity (EI) were measured in 12 healthy subjects and a myotonic dystrophy type 1 patient. RESULTS: MTs ranged from 0.15 to 0.30 mm, except for the procerus muscle (0.06 mm). EIs ranged from 1 to 34, except for the procerus muscle. MT reproducibility was fair for the orbicularis oris pars labialis, excellent for the procerus and levator labii, and good for the other muscles. The myotonic dystrophy type 1 patient showed high EIs, outside the range in healthy subjects in 6 of the 7 muscles. MT was lower than the range seen in healthy subjects in 4 muscles. CONCLUSION: Quantitative muscle ultrasound of the facial muscles is feasible and shows moderate to excellent reproducibility.