Cryoneurolysis of the Femoral Nerve for Focal Spasticity in an Ambulatory Patient.

Introduction: Spasticity of the knee extensors is a common presentation among patients with multiple sclerosis. The resulting stiff leg gait can result in increased risk of falls, heightened energy expenditure during gait, lowered gait speed, and compensatory gait mechanisms that increase wear on the hips. Cryoneurolysis is a novel percutaneous, minimally invasive treatment for focal spasticity. Methods: A single patient with multiple sclerosis was treated with cryoneurolysis of the femoral nerve branch to rectus femoris. The patient was followed for 15 months. Spasticity severity, gait speed, and patient reported outcomes were collected at each follow-up. Results: Spasticity severity as per the Modified Ashworth Scale was reduced at 1 month, with change persisting up to 15 months post-procedure. Range of motion as per the Modified Tardieu Scale showed gradual improvement over the 15-month period. Gait speed increased after the procedure from 21.15 seconds to 12.49 seconds for the 10 m walk test 1 month post-procedure, then slowed to baseline after 15 months. The patient's confidence in their gait improved and their independence was maintained throughout the follow-up period. Because of the regression in the 10 m walk test, the patient elected to have the procedure repeated after 15 months. Immediately after the procedure, the 10 m test time improved to 16.20 seconds. Conclusion: Cryoneurolysis of the femoral nerve may be an effective, long-lasting treatment for spasticity causing stiff knee gait in patients with multiple sclerosis.

Encoding the Latent Posterior of Bayesian Neural Networks for Uncertainty Quantification.

Bayesian Neural Networks (BNNs) have long been considered an ideal, yet unscalable solution for improving the robustness and the predictive uncertainty of deep neural networks. While they could capture more accurately the posterior distribution of the network parameters, most BNN approaches are either limited to small networks or rely on constraining assumptions, e.g., parameter independence. These drawbacks have enabled prominence of simple, but computationally heavy approaches such as Deep Ensembles, whose training and testing costs increase linearly with the number of networks. In this work we aim for efficient deep BNNs amenable to complex computer vision architectures, e.g., ResNet-50 DeepLabv3+, and tasks, e.g., semantic segmentation and image classification, with fewer assumptions on the parameters. We achieve this by leveraging variational autoencoders (VAEs) to learn the interaction and the latent distribution of the parameters at each network layer. Our approach, called Latent-Posterior BNN (LP-BNN), is compatible with the recent BatchEnsemble method, leading to highly efficient (in terms of computation and memory during both training and testing) ensembles. LP-BNNs attain competitive results across multiple metrics in several challenging benchmarks for image classification, semantic segmentation, and out-of-distribution detection.

A Case Report of Cryoneurolysis for Dorsal Foot Pain and Toe Clawing in a Patient With Multiple Sclerosis.

Toe clawing in patients with upper motor neuron disorders is often attributed to the flexor digitorum longus (FDL) and is a common presentation among patients with multiple sclerosis (MS). This movement may be painful because of the altered pressure distribution and may increase the risk of falls, heighten energy expenditure during gait, and lower gait speed. Cryoneurolysis is a minimally invasive treatment that may be beneficial for pain and focal muscle hypertonicity. An ambulatory patient with MS was treated bilaterally with cryoneurolysis to the superficial fibular nerves for pain on the dorsum of the foot, and to the intramuscular tibial nerve motor branch to FDL for toe clawing. The patient felt that toe clawing was immediately reduced during gait and noted the ability to voluntarily spread their toes. The patient stated that the neuropathic pain on the dorsum of the foot was fully eliminated immediately post procedure. The patient reported improved confidence in their gait, maintained independence, and reduced toe clawing during a structured interview 12 weeks after treatment. The effects lasted for 5.5 months before symptoms returned. Retreatment at 6 months reproduced the benefits. The patient reported a positive experience with cryoneurolysis for toe clawing and dorsal foot pain.