Cervical ganglioneuroma as a rare cause of cervicogenic headache: A case report and literature review.

BACKGROUND: Cervicogenic headache is characterized by chronic posterior neck pain radiating to one side of the head, resulting from cervical spine bone or soft tissue diseases. Cervical ganglioneuroma (GN), a rare benign neuroblastic tumor, especially in the cervical spine, may cause cervicogenic headache-like symptoms. OBJECTIVE: We report a case of GN which was surgically removed successfully to relieve the symptom. CASE REPORT: A 68-year-old male presented with right posterior neck pain with referred pain to the ipsilateral occipital area in May 2020. Despite administration of medications, physical therapy, and spine interventions, the symptoms intermittently recurred over one year. In July 2021, the patient complained of painful limitation of neck motion, especially on right-sided bending; no motor or sensory deficits, except for subjective numbness of the finger tips, were detected. Plain radiography of the cervical spine showed moderate degenerative changes in the mid-cervical spine. Cervical MRI revealed a cystic mass (1.5 cm × 0.5 cm × 1 cm-in size) around the right C2 dorsal root ganglion adjacent to the C1-C2 facet joint. His symptoms significantly improved after complete tumor excision. CONCLUSION: GN of the upper cervical spine should be considered when persistent cervicogenic headache is refractory to conservative management. In such a case, advanced imaging studies such as MRI should be performed for early diagnosis and appropriate treatment.

The application of a dual-lead locking screw could enhance the reduction and fixation stability of the proximal humerus fractures: a biomechanical evaluation.

Introduction: Bicortical screw fixation, which penetrates and fixes the near and far cortex of bone, has been conventionally used to achieve compressive fixation for fracture using screws. Open reduction and internal fixation using the locking plate are widely used for treating proximal humerus fractures. However, minimal contact between the bone and the locking plate can lead to an insufficient reduction. Theoretically, a dual-lead locking screw with different leads for the screw head and body could enhance the reduction and fixation stability of fragments in proximal humeral fractures without bicortical fixation, and achieve additional compression at the bone-plate-screw interface. This study assessed the insertion mechanics of the lead ratio of the dual-lead locking screw and its effect on the fixation stability of the proximal humerus fracture. Methods: A Multi-Fix® locking plating system composed of ∅ 3.5 mm locking screws and a locking plate was used to make a locked plating for Sawbone bone blocks and fourth-generation composite humeri. Two different types of Sawbone bone blocks were used to simulate the osteoporotic (10 PCF) and normal cancellous (20 PCF) bones. The lead of the screw head thread (Lhead) was 0.8 mm, and that of the screw body (Lbody) was 0.8, 1.25, 1.6, 2.0, and 2.4 mm, whose lead ratios (Rlead=Lbody/Lhead) were 1.0, 1.56, 2.0, 2.5, and 3.0, respectively. Results: The dual-lead locking screw elevated the compression between the locking plate and the bone. The elevation in the compression due to the dual-lead thread became weaker for the cancellous bone when the lead of the screw body was more than twice that of the screw head. The plate/humerus compression with strong bone quality withstood higher dual-lead-driven compression. Discussion: A dual-lead locking screw of Lbody=1.25mm (Rlead=1.56) is recommended for maximum rotational stability for the locked humerus plating. The screws with over Lbody=1.6mm (Rlead=2) have no advantage in terms of the failure torque and maximum torsional deformation. Any locking dual-lead screw with a body thread lead of <1.6 mm (Rlead=2) can be used without the risk of bone crush when surgeons require additional compression to the locked cancellous bone plating.

Artificial Octopus-Limb-Like Adhesive Patches for Cupping-Driven Transdermal Delivery with Nanoscale Control of Stratum Corneum.

Drug delivery through complex skin is currently being studied using various innovative structural and material strategies due to the low delivery efficiency of the multilayered stratum corneum as a barrier function. Existing microneedle-based or electrical stimulation methods have made considerable advances, but they still have technical limitations to reduce skin discomfort and increase user convenience. This work introduces the design, operation mechanism, and performance of noninvasive transdermal patch with dual-layered suction chamber cluster (d-SCC) mimicking octopus-limb capable of wet adhesion with enhanced adhesion hysteresis and physical stimulation. The d-SCC facilitates cupping-driven drug delivery through the skin with only finger pressure. Our device enables nanoscale deformation control of stratum corneum of the engaged skin, allowing for efficient transport of diverse drugs through the stratum corneum without causing skin discomfort. Compared without the cupping effect of d-SCC, applying negative pressure to the porcine, human cadaver, and artificial skin for 30 min significantly improved the penetration depth of liquid-formulated subnanoscale medicines up to 44, 56, and 139%. After removing the cups, an additional acceleration in delivery to the skin was observed. The feasibility of d-SCC was demonstrated in an atopic dermatitis-induced model with thickened stratum corneum, contributing to the normalization of immune response.

Soft tissue pseudotumors of the hand and wrist mimicking malignancy: Two case reports.

RATIONALE: Most soft-tissue lesions in the hand and wrist have a benign etiology, and malignant tumors, such as soft-tissue sarcomas, are rare. Mimickers of soft tissue tumors in the hand and wrist are more common than true neoplastic lesions; however, soft tissue pseudotumors that mimic malignancy are very rare. PATIENT CONCERNS: This study describes 2 patients with soft tissue pseudotumors of the hand and wrist. Both patients presented with rapidly growing soft-tissue masses. Magnetic resonance imaging (MRI) revealed ill-defined margins and an aggressive appearance in both cases, leading to a strong suspicion of malignant soft tissue tumors. DIAGNOSIS: Both patients underwent incisional biopsies, and the final diagnoses were inflammation due to IgG4-related disease in the first case and chronic granulomatous inflammation in the second case. INTERVENTIONS: The first patient was administered oral steroids, while the second patient was treated with anti-inflammatory drugs. OUTCOMES: Both patients showed a reduction in swelling of the hand and wrist. LESSONS: Although the imaging approach for pseudotumorous lesions is similar to that for true soft tissue tumors, the management of these lesions is different. Biopsies should only be performed when the diagnosis is unclear.

Concomitant tuberculosis of the iliotibial band mimicking a soft tissue tumor and tuberculous trochanteric bursitis: a case report emphasizing on magnetic resonance imaging findings.

The risk of tuberculosis (TB) increases in immunocompromised patients. Multiple myeloma is considered a risk factor for TB and myeloma patients with TB have a higher mortality rate than those without TB. Herein, we report a case of concomitant TB of the iliotibial band mimicking a soft tissue tumor and tuberculous trochanteric bursitis in a patient with multiple myeloma. In this article, the characteristic magnetic resonance imaging (MRI) findings were low T2 signals in the cystic fluid lesion, a dark T2 signal rim, and peripheral rim enhancement. These results could help differentiate TB of the iliotibial band and trochanteric bursitis from other pathologies. If the abovementioned findings were observed in immunocompromised patients, extrapulmonary TB may be expected even if chest radiographs are normal.

Average 22-Year Results of Total Hip Arthroplasty Using Harris-Galante Prosthesis in Patients under 50 Years.

Background: The Harris-Galante (HG) prosthesis is a first-generation, cementless total hip arthroplasty (THA) prosthesis. Considering the recent increase in the demand for THA in young patients and their life expectancy, a study with a follow-up duration of longer than 20 years in a young population is needed. Therefore, we evaluated the long-term clinical and radiographic results after cementless THA using the HG prosthesis in patients younger than 50 years. Methods: A total of 61 THAs performed using the HG with a minimum follow-up of 10 years were included. There were 38 men and 11 women with an average age of 46 years and the mean follow-up duration was 22 years. Clinical evaluation included modified Harris Hip Score (HHS) and radiographic analysis consisted of cup inclination, anteversion angle, component stability, osteolysis, liner wear rate, wear-through, liner dissociation, and heterotopic ossification. Complications included recurrent dislocation, periprosthetic femoral fracture, and periprosthetic joint infection. Survivorship analysis included cup and stem revision for aseptic loosening, as well as any revision. Results: The HHS improved from 46.5 preoperatively to 81.8 postoperatively (p < 0.001). The average linear wear rate was 0.36 mm/yr. A total of 34 hips (56%) were revised: stem revision in 10 (16.4%), cup revision in 9 (14.8%), exchange limited to bearing surface in 8 (13.1%), and revision of all components in 7 (11.5%). Estimated survivorship at 34 years postoperatively was 90.9% for cup revision for aseptic loosening, 80.5% for stem revision for aseptic loosening, and 22.1% for any revision. Conclusions: THA using the HG prosthesis showed satisfactory estimated survivorship of the acetabular and femoral components at 34 years postoperatively with good clinical outcomes. Bearing-related problems, such as osteolysis and liner dissociation, accounted for 56% of revision operations and were concerns in patients younger than 50 years.

Nerve root anomaly as a cause of unexpected outcomes of epidural steroid injections in lumbar radiculopathy: A case report.

BACKGROUND: Spinal nerve root anomaly is a rare feature that can result in unexpected outcomes in epidural steroid injections or surgical procedures. Preoperative diagnostic tools for root anomalies are limited, as they are usually found intraoperatively. OBJECTIVE: This case report aims to propose an effective diagnostic process for nerve root anomalies by introducing clinical manifestations, electrodiagnostic findings, and sophisticated imaging techniques such as coronal view magnetic resonance imaging (MRI) of the lumbosacral spine. CASE DESCRIPTION: A 43-year-old female complained of low back pain with radicular pain to the lower extremities. Based on physical examination, electrodiagnosis, and imaging studies, right L5 radiculopathy was diagnosed. Repetitive image-guided epidural steroid injections presented unsuccessful outcomes. She was then referred to a neurosurgeon for surgical decompression, which resulted in significant improvement in her radicular pain. A nerve root anomaly was found intraoperatively, and the coronal images of postoperative MRI depicted the conjoined nerve root of the lumbar spine. CONCLUSION: When proper image-guided spinal interventions for discogenic radicular pain are not effective, a clinician should be advised to consider the possibility of anatomical variation, including nerve root anomalies. Early diagnosis of nerve root anomalies by utilizing multiple diagnostic tools, especially coronal MRI, can aid in preoperative diagnosis and proper clinical decisions for symptom management.

COVID-19 Detection in Chest X-ray Images Using a New Channel Boosted CNN.

COVID-19 is a respiratory illness that has affected a large population worldwide and continues to have devastating consequences. It is imperative to detect COVID-19 at the earliest opportunity to limit the span of infection. In this work, we developed a new CNN architecture STM-RENet to interpret the radiographic patterns from X-ray images. The proposed STM-RENet is a block-based CNN that employs the idea of split-transform-merge in a new way. In this regard, we have proposed a new convolutional block STM that implements the region and edge-based operations separately, as well as jointly. The systematic use of region and edge implementations in combination with convolutional operations helps in exploring region homogeneity, intensity inhomogeneity, and boundary-defining features. The learning capacity of STM-RENet is further enhanced by developing a new CB-STM-RENet that exploits channel boosting and learns textural variations to effectively screen the X-ray images of COVID-19 infection. The idea of channel boosting is exploited by generating auxiliary channels from the two additional CNNs using Transfer Learning, which are then concatenated to the original channels of the proposed STM-RENet. A significant performance improvement is shown by the proposed CB-STM-RENet in comparison to the standard CNNs on three datasets, especially on the stringent CoV-NonCoV-15k dataset. The good detection rate (97%), accuracy (96.53%), and reasonable F-score (95%) of the proposed technique suggest that it can be adapted to detect COVID-19 infected patients.

Electrostatic-Mechanical Synergistic In Situ Multiscale Tissue Adhesion for Sustainable Residue-Free Bioelectronics Interfaces.

Recent studies on soft adhesives have sought to deeply understand how their chemical or mechanical structures interact strongly with living tissues. The aim is to optimally address the unmet needs of patients with acute or chronic diseases. Synergistic adhesion involving both electrostatic (hydrogen bonds) and mechanical interactions (capillarity-assisted suction stress) seems to be effective in overcoming the challenges associated with long-term unstable coupling to tissues. Here, an electrostatically and mechanically synergistic mechanism of residue-free, sustainable, in situ tissue adhesion by implementing hybrid multiscale architectonics. To deduce the mechanism, a thermodynamic model based on a tailored multiscale combinatory adhesive is proposed. The model supports the experimental results that the thermodynamically controlled swelling of the nanoporous hydrogel embedded in the hierarchical elastomeric structure enhances biofluid-insensitive, sustainable, in situ adhesion to diverse soft, slippery, and wet organ surfaces, as well as clean detachment in the peeling direction. Based on the robust tissue adhesion capability, universal reliable measurements of electrophysiological signals generated by various tissues, ranging from rodent sciatic nerve, the muscle, brain, and human skin, are successfully demonstrated.

Deep vein thrombosis in the thigh: MR imaging in two cases with atypical presentations.

Deep vein thrombosis (DVT) is a common clinical problem affecting the lower extremities. Prompt imaging of suspected DVT is helpful for rapid diagnosis and proper treatment. However, patients without clear predisposing factors for DVT may be directed to alternative diagnoses of a musculoskeletal disorder. The few case reports and studies of magnetic resonance (MR) imaging of unsuspected DVT are limited to the calf and knee. Here, we report two cases with a rare presentation of thigh MR imaging of unsuspected DVT. Identifying branching, abnormal intraluminal signals on fluid-sensitive imaging, or rim-enhancing tubular structures within the edema of the thigh muscle is important for differentiating intramuscular DVT from other thigh pathologies.

Detection of tumour infiltrating lymphocytes in CD3 and CD8 stained histopathological images using a two-phase deep CNN.

BACKGROUND: Immuno-score, a prognostic measure for cancer, employed in determining tumor grade and type, is generated by counting the number of Tumour-Infiltrating Lymphocytes (TILs) in CD3 and CD8 stained histopathological tissue samples. Significant stain variations and heterogeneity in lymphocytes' spatial distribution and density make automated counting of TILs' a challenging task. METHODS: This work addresses the aforementioned challenges by developing a pipeline "Two-Phase Deep Convolutional Neural Network based Lymphocyte Counter (TDC-LC)" to detect lymphocytes in CD3 and CD8 stained histology images. The proposed pipeline sequentially works by removing hard negative examples (artifacts) in the first phase using a custom CNN "LSATM-Net" that exploits the idea of a split, asymmetric transform, and merge. Whereas, in the second phase, instance segmentation is performed to detect and generate a lymphocyte count against the remaining samples. Furthermore, the effectiveness of the proposed pipeline is measured by comparing it with the state-of-the-art single- and two-stage detectors. The inference code is available at GitHub Repository https://github.com/m-mohsin-zafar/tdc-lc. RESULTS: The empirical evaluation on samples from LYSTO dataset shows that the proposed LSTAM-Net can learn variations in the images and precisely remove the hard negative stain artifacts with an F-score of 0.74. The detection analysis shows that the proposed TDC-LC outperforms the existing models in identifying and counting lymphocytes with high Recall (0.87) and F-score (0.89). Moreover, the commendable performance of the proposed TDC-LC in different organs suggests a good generalization. CONCLUSION: The promising performance of the proposed pipeline suggests that it can serve as an automated system for detecting and counting lymphocytes from patches of tissue samples thereby reducing the burden on pathologists.

COVID-19 detection in chest X-ray images using deep boosted hybrid learning.

The new emerging COVID-19, declared a pandemic disease, has affected millions of human lives and caused a massive burden on healthcare centers. Therefore, a quick, accurate, and low-cost computer-based tool is required to timely detect and treat COVID-19 patients. In this work, two new deep learning frameworks: Deep Hybrid Learning (DHL) and Deep Boosted Hybrid Learning (DBHL), is proposed for effective COVID-19 detection in X-ray dataset. In the proposed DHL framework, the representation learning ability of the two developed COVID-RENet-1 & 2 models is exploited individually through a machine learning (ML) classifier. In COVID-RENet models, Region and Edge-based operations are carefully applied to learn region homogeneity and extract boundaries features. While in the case of the proposed DBHL framework, COVID-RENet-1 & 2 are fine-tuned using transfer learning on the chest X-rays. Furthermore, deep feature spaces are generated from the penultimate layers of the two models and then concatenated to get a single enriched boosted feature space. A conventional ML classifier exploits the enriched feature space to achieve better COVID-19 detection performance. The proposed COVID-19 detection frameworks are evaluated on radiologist's authenticated chest X-ray data, and their performance is compared with the well-established CNNs. It is observed through experiments that the proposed DBHL framework, which merges the two-deep CNN feature spaces, yields good performance (accuracy: 98.53%, sensitivity: 0.99, F-score: 0.98, and precision: 0.98). Furthermore, a web-based interface is developed, which takes only 5-10s to detect COVID-19 in each unseen chest X-ray image. This web-predictor is expected to help early diagnosis, save precious lives, and thus positively impact society.

Intramuscular peripheral nerve sheath tumors: schwannoma, ancient schwannoma, and neurofibroma.

PURPOSE: To analyze the prevalence of classic magnetic resonance imaging (MRI) findings of intramuscular peripheral nerve sheath tumors (PNSTs), including schwannoma, ancient schwannoma, and neurofibroma. METHOD: Thirty pathologically confirmed benign intramuscular PNSTs (24 schwannomas, 3 ancient schwannomas, and 3 neurofibromas) were retrospectively reviewed. Classic MRI findings of PNSTs including split fat sign, fascicular sign, target sign, entering and exiting nerve, and thin hyperintense rim were assessed for each intramuscular PNST. Denervation change of the affected muscle was also assessed. In ancient schwannoma and neurofibroma, the signal intensity (SI) and enhancement pattern were analyzed. RESULTS: All intramuscular schwannomas revealed two more classic MRI findings. Eight of the 24 intramuscular schwannomas revealed affected muscle denervation change. All intramuscular ancient schwannomas showed only split fat sign. All intramuscular ancient schwannomas showed denervation change of the associated muscle. All intramuscular neurofibroma showed split fat sign and one case with target sign was detected. Ancient schwannomas were isointense SI on T1-weighted image (T1WI) and one case had hyperintense foci. They showed heterogeneously hyperintense SI on T2-weighted image (T2WI) with heterogeneous enhancement. Neurofibromas were isointense SI (2/3) and slight hyperintense SI (1/3) on T1WI and heterogeneously hyperintense SI on T2WI with heterogeneous enhancement. One ancient schwannoma showed conglomerated calcifications. CONCLUSIONS: Intramuscular schwannomas were easily diagnosed based on MRI. In the case of intramuscular ancient schwannoma and neurofibroma with only split fat sign among the classic MRI findings, they might be distinguished from other intramuscular soft tissue tumors based on muscle denervation change or typical crescent split fat sign.

A rare case of multiple and huge simple bone cysts in Wilson's disease.

A simple bone cyst is a tumor-like lesion that is not a true neoplasm. It usually presents as a solitary lesion within the proximal humerus and proximal femur. The etiology of simple bone cysts is still to be elucidated. We describe our experience in the successful surgical management of a rare case of two, large-sized, simple bone cysts involving the entire left humerus and right femoral head, respectively, in a 34-year-old male patient with Wilson's disease and concomitant hepatic cirrhosis. We theorized that Wilson's disease could be the cause of the delay or blockage of the natural healing process of the simple long-bone cysts, leading to their abnormal growth and development. To our knowledge, such a combined occurrence of multiple, simple long-bone cysts and Wilson's disease has not been reported in the literature to date.

Transfer learning based deep CNN for segmentation and detection of mitoses in breast cancer histopathological images.

Segmentation and detection of mitotic nuclei is a challenging task. To address this problem, a Transfer Learning based fast and accurate system is proposed. To give the classifier a balanced dataset, this work exploits the concept of Transfer Learning by first using a pre-trained convolutional neural network (CNN) for segmentation, and then another Hybrid-CNN (with Weights Transfer and custom layers) for classification of mitoses. First, mitotic nuclei are automatically annotated, based on the ground truth centroids. The segmentation module then segments mitotic nuclei and also produces some false positives. Finally, the detection module is trained on the patches from the segmentation module and performs the final detection. Fine-tuning based Transfer Learning reduced training time, provided good initial weights, and improved the detection rate with F-measure of 0.713 and 76% area under the precision-recall curve for the challenging task of mitosis detection.

NKX6.3 protects against gastric mucosal atrophy by downregulating ß-amyloid production.

BACKGROUND: Atrophic gastritis is characterized by loss of appropriate glands and reduction in gastric secretory function due to chronic inflammatory processes in gastric mucosa. Moreover, atrophic gastritis is considered as a precancerous condition of gastric cancer. However, little is known about the molecular mechanism underlying gastric mucosal atrophy and its contribution to gastric carcinogenesis. Thus, we hypothesized that transcription factor NKX6.3 might be involved in maintaining gastric epithelial homeostasis by regulating amyloid ß (Aß) production. AIM: To determine whether NKX6.3 might protect against gastric mucosal atrophy by regulating Aß production. METHODS: We identified NKX6.3 depletion induced cell death by cell count and Western blot assay. Production and mechanism of Aß oligomer were analyzed by enzyme-linked immunosorbent assay, Western blot, immunoprecipitation, real-time quantitative polymerase chain reaction and immunofluorescence analysis. We further validated the correlation between expression of NKX6.3, Helicobacter pylori CagA, Aß oligomer, apolipoprotein E (ApoE), and ß-secretase 1 (Bace1) in 55 gastric mucosae. RESULTS: NKX6.3 depletion increased both adherent and floating cell populations in HFE-145 cells. Expression levels of cleaved caspase-3, -9, and poly ADP ribose polymerase were elevated in floating HFE-145shNKX6.3 cells. NKX6.3 depletion produced Aß peptide oligomers, and increased expression of ApoE, amyloid precursor protein, Aß, Bace1, low-density lipoprotein receptor, nicastrin, high mobility group box1, and receptor for advanced glycosylation end product proteins. In immunoprecipitation assay, γ-secretase complex was stably formed only in HFE-145shNKX6.3 cells. In gastric mucosae with atrophy, expression of Aß peptide oligomer, ApoE, and Bace1 was detected and inversely correlated with NKX6.3 expression. Treatment with recombinant Aß 1-42 produced Aß oligomeric forms and decreased cell viability in HFE-145shNKX6.3 cells. Additionally, NKX6.3 depletion increased expression of inflammatory cytokines and cyclooxygenase-2. CONCLUSION: NKX6.3 inhibits gastric mucosal atrophy by regulating Aß accumulation and inflammatory reaction in gastric epithelial cells.

Changes in ankle joint motion after Supramalleolar osteotomy: a cadaveric model.

BACKGROUND: Malalignment of the ankle joint has been found after trauma, by neurological disorders, genetic predisposition and other unidentified factors, and results in asymmetrical joint loading. For a medial open wedge supramalleolar osteotomy(SMO), there are some debates as to whether concurrent fibular osteotomy should be performed. We assessed the changes in motion of ankle joint and plantar pressure after supramalleolar osteotomy without fibular osteotomy. METHODS: Ten lower leg specimens below the knee were prepared from fresh-frozen human cadavers. They were harvested from five males (10 ankles)whose average age was 70 years. We assessed the motion of ankle joint as well as plantar pressure for SS(supra-syndesmotic) SMO and IS(intra-syndesmotic) SMO. After the osteotomy, each specimen was subjected to axial compression from 20 N preload to 350 N representing half-body weight. For the measurement of the motion of ankle joint, the changes in gap and point, angles in ankle joint were measured. The plantar pressure were also recorded using TekScan sensors. RESULTS: The changes in the various gap, point, and angles movements on SS-SMO and IS-SMO showed no statistically significant differences between the two groups. Regarding the shift of plantar center of force (COF) were noted in the anterolateral direction, but not statistically significant. CONCLUSIONS: SS-SMO and IS-SMO with intact fibula showed similar biomechanical effect on the ankle joint. We propose that IS-SMO should be considered carefully for the treatment of osteoarthrosis when fibular osteotomy is not performed because lateral cortex fracture was less likely using the intrasyndesmosis plane because of soft tissue support.

Ultrasonographic changes after steroid injection in carpal tunnel syndrome.

OBJECTIVE: To determine the ultrasonographic changes after steroid injection in carpal tunnel syndrome (CTS) and to evaluate the diagnostic value of ultrasound in post-treatment examination with clinical correlation. MATERIALS AND METHODS: Twenty-seven wrists with idiopathic CTS after a single injection of 40 mg of prednisolone hydrochloride were prospectively studied using a high-resolution ultrasound. Axial images of the wrists were obtained at the level of the distal radius, pisiform and hamate prior to and 1, 4 and 8 weeks after steroid injection. The cross-sectional area (CSA, mm2) and flattening ratio (FR) of the median nerve were measured. The bowing of the flexor retinaculum (palmar displacement: PD, mm) and the transverse sliding distance of the median nerve (TSD, mm) during flexion-extension of the index finger were computed. Pre- and post-injection ultrasonographic findings were analyzed in relation to clinical parameters such as pain score. RESULTS: At all levels of the wrist, the CSA decreased significantly at 1 week after the injection, which continued to 8 weeks post-injection. The PD also diminished significantly 1 week after the injection. Furthermore, the TSD significantly increased from 1 week after injection, which lasted to the end of this study. The CSA, PD and TSD showed significant correlation with improvement of pain scores. CONCLUSION: Ultrasound is useful in follow-up examinations of CTS. Significant ultrasound findings after steroid injections in CTS include decreased swelling of the median nerve, decreased bowing of the flexor retinaculum and increased mobility of the median nerve. All of them significantly correlate with clinical symptoms.

Two-phase deep convolutional neural network for reducing class skewness in histopathological images based breast cancer detection.

Different types of breast cancer are affecting lives of women across the world. Common types include Ductal carcinoma in situ (DCIS), Invasive ductal carcinoma (IDC), Tubular carcinoma, Medullary carcinoma, and Invasive lobular carcinoma (ILC). While detecting cancer, one important factor is mitotic count - showing how rapidly the cells are dividing. But the class imbalance problem, due to the small number of mitotic nuclei in comparison to the overwhelming number of non-mitotic nuclei, affects the performance of classification models. This work presents a two-phase model to mitigate the class biasness issue while classifying mitotic and non-mitotic nuclei in breast cancer histopathology images through a deep convolutional neural network (CNN). First, nuclei are segmented out using blue ratio and global binary thresholding. In Phase-1 a CNN is then trained on the segmented out 80×80 pixel patches based on a standard dataset. Hard non-mitotic examples are identified and augmented; mitotic examples are oversampled by rotation and flipping; whereas non-mitotic examples are undersampled by blue ratio histogram based k-means clustering. Based on this information from Phase-1, the dataset is modified for Phase-2 in order to reduce the effects of class imbalance. The proposed CNN architecture and data balancing technique yielded an F-measure of 0.79, and outperformed all the methods relying on specific handcrafted features, as well as those using a combination of handcrafted and CNN-generated features.