Multi-Dimensional Wi-Fi Received Signal Strength Indicator Data Augmentation Based on Multi-Output Gaussian Process for Large-Scale Indoor Localization.

Location fingerprinting using Received Signal Strength Indicators (RSSIs) has become a popular technique for indoor localization due to its use of existing Wi-Fi infrastructure and Wi-Fi-enabled devices. Artificial intelligence/machine learning techniques such as Deep Neural Networks (DNNs) have been adopted to make location fingerprinting more accurate and reliable for large-scale indoor localization applications. However, the success of DNNs for indoor localization depends on the availability of a large amount of pre-processed and labeled data for training, the collection of which could be time-consuming in large-scale indoor environments and even challenging during a pandemic situation like COVID-19. To address these issues in data collection, we investigate multi-dimensional RSSI data augmentation based on the Multi-Output Gaussian Process (MOGP), which, unlike the Single-Output Gaussian Process (SOGP), can exploit the correlation among the RSSIs from multiple access points in a single floor, neighboring floors, or a single building by collectively processing them. The feasibility of MOGP-based multi-dimensional RSSI data augmentation is demonstrated through experiments using the hierarchical indoor localization model based on a Recurrent Neural Network (RNN)-i.e., one of the state-of-the-art multi-building and multi-floor localization models-and the publicly available UJIIndoorLoc multi-building and multi-floor indoor localization database. The RNN model trained with the UJIIndoorLoc database augmented with the augmentation mode of "by a single building", where an MOGP model is fitted based on the entire RSSI data of a building, outperforms the other two augmentation modes and results in the three-dimensional localization error of 8.42 m.

Safety of Outpatient Chest Tube Management of Air Leaks After Pulmonary Resection.

Prolonged air leaks are the most common postoperative complication following pulmonary resection, leading to increased hospital length of stay (LOS) and cost. This study assesses the safety of discharging patients home with a chest tube (CT) after pulmonary resection. A retrospective review was performed of a single surgeon's experience with pulmonary resections from January 2010 to January 2015. All patients discharged home with a CT were included. Discharge criteria included a persistent air leak controlled by water seal, resolution of medical conditions requiring hospitalization, and pain managed by oral analgesics. Patient demographics, type of resection, LOS, and 30-day morbidity and mortality data were analyzed. Comparisons were made with the Society of Thoracic Surgery database January 2011 to December 2013. Four hundred ninety-six patients underwent pulmonary resection. Sixty-five patients (13%) were discharged home postoperatively with a CT. Fifty-eight patients underwent a lobectomy, two patients a bilobectomy, and five patients had a wedge excision. Two patients were readmitted: One with a lower extremity deep venous thrombosis and the other with a nonlife threatening pulmonary embolus. Four patients developed superficial CT site infections that resolved after oral antibiotics. Patients discharged home with a CT following lobectomy had a shorter mean LOS compared to lobectomy patients (3.65 vs 6.2 days). Mean time to CT removal after discharge was 4.7 days (range 1-22 days) potentially saving 305 inpatient hospital days. Select patients can be discharged home with a CT with reduced postoperative LOS and without increase in major morbidity or mortality.

Cervical spine surgery: anterior microsurgery.

Anterior cervical decompression and fusion is one of the most commonly used procedures to treat myelopathy and radiculopathy. With the addition of the surgical microscope, enhanced visualization of the microanatomy has allowed the safe and efficient treatment of compressive pathology. It is important to understand the critical steps required to successfully treat cervical spine pathology from an anterior approach with the assistance of a surgical microscope, the factors involved in choosing a decompressive method, the technical considerations for intervertebral grafting, and the use of cervical plating. The knowledge of surgical pearls may optimize spinal cord and nerve root decompression, increase fusion rates, and give patients the best chance for a successful clinical outcome.

Role of early surgical decompression of the intradural space after cervical spinal cord injury in an animal model.

BACKGROUND: The role of decompressing the intradural space through a durotomy as a treatment option for acute traumatic cervical spinal cord injury has not been explored in an animal model, to our knowledge. We sought to determine the role of durotomy and duraplasty in the treatment of acute cervical spinal cord injury and its effects on inflammation, scar formation, and functional recovery. METHODS: Seventy-two adult female Sprague-Dawley rats were assigned to three groups: contusion injury alone, contusion injury with a decompressive durotomy, and contusion injury with a decompressive durotomy followed by placement of a dural allograft. A mild (200-kdyn [2-N]) contusive injury was delivered to the exposed spinal cord at C5. The injured segment was reexposed four hours after injury, and a durotomy with decompression was performed. When a dural allograft was used it was affixed to the surrounding intact dura with use of a fibrin sealant. The Grip Strength Meter was used to assess forelimb function. Animals were killed at two and four weeks, and immunohistochemical analysis was performed to assess scar formation, inflammatory cell infiltration, and lesional volume. RESULTS: Immunohistochemical analysis revealed increased scar formation, cavitation, and inflammatory response in the animals treated only with a decompressive durotomy. Relative to the group with a contusion injury alone, the animals treated with a durotomy followed by a dural allograft had decreased cavitation and scar formation. Lesional volume measurements showed a significantly increased cavitation size at four weeks in both the contusion-only (mean and standard deviation, 12.6 +/- 0.5 mm(3)) and durotomy-only (15.1 +/- 1 mm(3)) groups relative to the animals that had received a dural allograft following durotomy (6.8 +/- 1.4 mm(3)). CONCLUSIONS: Functional recovery after acute cervical spinal cord injury was better in animals treated with decompression of the intradural space and placement of a dural allograft than it was in animals treated with decompression alone. These functional data correlated directly with histological evidence of a decrease in spinal cord cavitation, inflammation, and scar formation.

Advances in the management of spinal cord injury.

Historically, clinical outcomes following spinal cord injury have been dismal. Over the past 20 years, the survival rate and long-term outcome of patients with spinal cord injury have improved with advances in both medical and surgical treatment. However, the efficacy and timing of these adjuvant treatments remain controversial. There has been a tremendous increase in the number of basic science and clinical studies on spinal cord injury. Current areas of investigation include early acute management, including early surgical intervention, as well as new pharmacotherapy and cellular transplantation strategies. It is unlikely that a single approach can uniformly address all of the issues associated with spinal cord injury. Thus, a multidisciplinary approach will be needed.

Sustained spinal cord compression: part II: effect of methylprednisolone on regional blood flow and recovery of somatosensory evoked potentials.

BACKGROUND: The efficacy of methylprednisolone in the treatment of traumatic spinal cord injury is controversial. We examined the effect of methylprednisolone on regional spinal cord blood flow and attempted to determine whether recovery of electrophysiological function is dependent on reperfusion, either during sustained spinal cord compression or after decompression. METHODS: The effects of methylprednisolone therapy on recovery of somatosensory evoked potentials and on spinal cord blood flow were examined in a canine model of dynamic spinal cord compression. Methylprednisolone (30 mg/kg intravenous loading dose followed by 5.4 mg/kg/hr intravenous infusion) or saline solution was administered to thirty-six beagles (eighteen in each group) five minutes after cessation of dynamic spinal cord compression and loss of all somatosensory evoked potentials. After ninety minutes of sustained compression, the spinal cords were decompressed. Somatosensory evoked potentials and spinal cord blood flow were evaluated throughout the period of sustained compression and for three hours after decompression. RESULTS: Seven dogs treated with methylprednisolone and none treated with saline solution recovered measurable somatosensory evoked potentials during sustained compression. After decompression, three more dogs treated with methylprednisolone and seven dogs treated with saline solution recovered somatosensory evoked potentials. Four dogs treated with methylprednisolone lost their previously measurable somatosensory evoked potentials. In the methylprednisolone group, spinal cord blood flow was significantly higher (p < 0.05) in the dogs that had recovered somatosensory evoked potentials than it was in the dogs that had not. Reperfusion blood flow was significantly higher (p < 0.05) in the saline-solution group than it was in the methylprednisolone group. Spinal cord blood flow in the saline-solution group returned to baseline levels within five minutes after decompression. It did not return to baseline levels in the dogs treated with methylprednisolone. CONCLUSIONS: The methylprednisolone administered in this study did not provide a large or significant lasting benefit with regard to neurological preservation or restoration. Methylprednisolone may reduce regional spinal cord blood flow through mechanisms affecting normal autoregulatory blood-flow function.