A deep learning model to triage and predict adenocarcinoma on pancreas cytology whole slide imaging.

Pancreatic fine-needle aspirations are the gold-standard diagnostic procedure for the evaluation of pancreatic ductal adenocarcinoma. A suspicion for malignancy can escalate towards chemotherapy followed by a major surgery and therefore is a high-stakes task for the pathologist. In this paper, we propose a deep learning framework, MIPCL, that can serve as a helpful screening tool, predicting the presence or absence of cancer. We also reproduce two deep learning models that have found success in surgical pathology for our cytopathology study. Our MIPCL significantly improves over both models across all evaluated metrics (F1-Score: 87.97% vs 88.70% vs 91.07%; AUROC: 0.9159 vs. 0.9051 vs 0.9435). Additionally, our model is able to recover the most contributing regions on the slide for the final prediction. We also present a dataset curation strategy that increases the number of training examples from an existing dataset, thereby reducing the resource burden tied to collecting and scanning additional cases.

De Quervain Tenosynovitis: An Evaluation of the Epidemiology and Utility of Multiple Injections Using a National Database.

PURPOSE: We hypothesized that repeat injections are associated with a decreased rate of success and that the success rate of injections correlates with patient comorbidities. METHODS: Using a commercially available insurance database, patients diagnosed with De Quervain tenosynovitis were identified using International Classification of Diseases, Ninth Revision and Tenth Revision codes and stratified by therapeutic interventions, including therapy, injections, and surgery, as well as comorbidities. Injection failure was defined as a patient receiving a repeat injection or subsequent surgical management. Success was defined as no further therapies identified after an intervention. RESULTS: From 2007 to 2017, 33,420 patients with a primary diagnosis of De Quervain tenosynovitis were identified. Women represented 77.5% (25,908) of the total and were 2.6 times more likely to be diagnosed than men. Black patients were more likely to be diagnosed than White patients. Black and White women were found to have the highest incidence (relative risk 3.4 and 2.3, respectively, compared with White men). Age was also significantly correlated with an increased risk of diagnosis of the condition, with a peak incidence at the age of 40-59 years (relative risk, 10.6). Diabetes, rheumatoid arthritis, lupus, and hypothyroidism were associated with an increased risk of diagnosis. Overall, 53.3% of the patients were treated with injections, 11.6% underwent surgery, and 5.2% underwent therapy. Treatment with a single injection was successful in 71.9% of the patients, with 19.7% receiving a repeat injection and 8.4% treated with surgery. The overall success rate of subsequent injections was 66.3% for the second injection and 60.5% for the third. The initial injection had a higher rate of success in diabetics than in nondiabetics; however, the difference (2%) was not clinically relevant. CONCLUSIONS: Although the success rate for the treatment of De Quervains tenosynovitis decreases with multiple injections, repeat injections have a high rate of success and are a viable clinical option. TYPE OF STUDY/LEVEL OF EVIDENCE: Therapeutic II.

Chronic Preoperative Opioids Are Associated With Revision After Rotator Cuff Repair.

PURPOSE: We sought to clarify the relationship between chronic preoperative opioids and complications following rotator cuff repair. Specifically, we assessed revision, a definitive postoperative end point for surgical outcome. METHODS: This study used PearlDiver, a United States national insurance claims database. All patients undergoing rotator cuff repair from 2008 to 2018 were identified and stratified based on a minimum of 2 opioid prescriptions within the 6 months before surgery, with 1 prescription occurring within 0 to 3 months before surgery and a second prescription within 4 to 6 months before surgery. Univariate logistic regressions of risk factors were conducted, followed by multivariate analysis of comorbidities, including ongoing preoperative opioids, any preoperative nonsteroidal anti-inflammatory drug (NSAID) prescriptions, age, sex, diabetes, tobacco, and obesity. RESULTS: In total, 28,939 patients undergoing rotator cuff repair were identified, of whom 10,695 had opioid prescriptions within both 0 to 3 months and 4 to 6 months before index rotator cuff repair, whereas 18,244 had no opioid prescriptions within the 6-month preoperative period. In total, 977 (3.4%) patients underwent revision within 6 months, which increased to 1311 (4.5%) within 1 year of the index procedure. In the multivariate analysis controlling for age, preoperative NSAID prescriptions, tobacco, diabetes, obesity, and sex, we observed a significant association between chronic preoperative opioid prescriptions and rotator cuff repair revision (6-month odds ratio 1.12; P = .021, 1-year odds ratio 1.43; P < .001) following index procedure. CONCLUSIONS: We report increased rates of revision within both 6 months and 1 year in patients with prolonged preoperative opioid prescriptions. The opioid cohort had greater rates of preoperative NSAID use and tobacco use, which also were observed to be independent risk factors for revision at both timepoints. LEVEL OF EVIDENCE: III; Retrospective comparative study.

Transcoracoacromial Ligament Glenohumeral Injection Technique Using Bony Surface Landmarks.

Intra-articular glenohumeral injection is an important technique used to diagnose and treat shoulder disorders. However, it is frequently performed as an image-guided technique with the use of fluoroscopy, ultrasound, computed tomography, or magnetic resonance. The purpose of this Technical Note is to describe a transcoracoacromial ligament glenohumeral injection technique that uses anatomic surface landmarks to avoid the need for radiographic guidance. After identification of the anterolateral corner of acromion, the superior lateral border of the coracoid tip, and the curved depression of the distal clavicle, the needle entry site is determined at the trisection point between the distal and middle thirds of the line formed by the superior lateral border of the coracoid tip and the curved depression of the distal clavicle. The needle is first inserted perpendicular to the triangular plane of the 3 points and is then advanced toward the humeral head. This injection technique is highly accurate and reproducible and can be done in the outpatient clinic without the use of imaging guidance, reducing the costs and barriers of intra-articular glenohumeral injections for patients.

Spatial aggregation of holistically-nested convolutional neural networks for automated pancreas localization and segmentation.

Accurate and automatic organ segmentation from 3D radiological scans is an important yet challenging problem for medical image analysis. Specifically, as a small, soft, and flexible abdominal organ, the pancreas demonstrates very high inter-patient anatomical variability in both its shape and volume. This inhibits traditional automated segmentation methods from achieving high accuracies, especially compared to the performance obtained for other organs, such as the liver, heart or kidneys. To fill this gap, we present an automated system from 3D computed tomography (CT) volumes that is based on a two-stage cascaded approach-pancreas localization and pancreas segmentation. For the first step, we localize the pancreas from the entire 3D CT scan, providing a reliable bounding box for the more refined segmentation step. We introduce a fully deep-learning approach, based on an efficient application of holistically-nested convolutional networks (HNNs) on the three orthogonal axial, sagittal, and coronal views. The resulting HNN per-pixel probability maps are then fused using pooling to reliably produce a 3D bounding box of the pancreas that maximizes the recall. We show that our introduced localizer compares favorably to both a conventional non-deep-learning method and a recent hybrid approach based on spatial aggregation of superpixels using random forest classification. The second, segmentation, phase operates within the computed bounding box and integrates semantic mid-level cues of deeply-learned organ interior and boundary maps, obtained by two additional and separate realizations of HNNs. By integrating these two mid-level cues, our method is capable of generating boundary-preserving pixel-wise class label maps that result in the final pancreas segmentation. Quantitative evaluation is performed on a publicly available dataset of 82 patient CT scans using 4-fold cross-validation (CV). We achieve a (mean  ±â€¯ std. dev.) Dice similarity coefficient (DSC) of 81.27 ±â€¯6.27% in validation, which significantly outperforms both a previous state-of-the art method and a preliminary version of this work that report DSCs of 71.80 ±â€¯10.70% and 78.01 ±â€¯8.20%, respectively, using the same dataset.

Infrared camera evaluation of the cooling effect of triamcinolone acetonide aerosol.

OBJECTIVE: Triamcinolone acetonide spray is a topical corticosteroid indicated for the relief of inflammatory/pruritic manifestations of corticosteroid-responsive dermatoses. There are clinical reports of an antipruritic, cooling sensation appreciated upon application. This study was designed to quantify the cryotherapeutic cooling effect of triamcinolone acetonide spray. DESIGN: Using an infrared video camera, skin surface temperature was evaluated for change upon application of the triamcinolone acetonide and two comparator ingredient components of triamcinolone acetonide: ethanol alcohol in a non-aerosolized spray and triamcinolone acetonide cream. SETTING: This was an open-label, single center, comparator study. PARTICIPANTS: This study enrolled 20 subjects with a diagnosis of either an acute or chronic steroid-responsive dermatosis. Ten additional controls were also enrolled. MEASUREMENTS: Using an infrared video camera, skin surface temperature was evaluated for change upon application of the triamcinolone acetonide and two comparator ingredient components of triamcinolone acetonide:ethanol alcohol in a non-aerosolized spray and triamcinolone acetonide cream. RESULTS: Across every study cohort, the average change in skin surface temperature with triamcinolone acetonide (between 16-18°C; P<0.001 for all comparisons, Figures 1 and 2) was significantly greater than the change demonstrated by both the non-aerosolized spray (between 5-7°C) and the triamcinolone acetonide cream (between 5.0-6.5°C). CONCLUSION: The transient temperature change of nearly 20°C with triamcinolone acetonide is most likely attributable to the refrigerant properties of the isobutane propellant of this product. Similar to other common cryotherapy methods, triamcinolone acetonide can achieve very low skin surface temperatures, which may result in localized relief of pruritus.

Eczema.

Atopic dermatitis, commonly known as eczema, is a common chronic, relapsing skin disease characterized by pruritus, disrupted epidermal barrier function, and immunoglobulin E-mediated sensitization to food and environmental allergens. Atopic dermatitis is a complex disease that arises from interactions between genes and the environment. Loci on several chromosomes have been identified, including a family of epithelium-related genes called the epidermal differentiation complex on chromosome 1q21. Mutations in filaggrin, a key protein in epidermal differentiation, have also been identified in early-onset and severe atopic dermatitis. There are 3 classical stages of eczema: infantile, childhood, and adulthood. The spectrum of eczema presentation varies widely from a variant that only affect the hand to major forms where a patient presents with erythroderma. The acute and subacute lesions of atopic dermatitis are often characterized by intensely pruritic, erythematous papules and vesicles with excoriations and a serous exudate. Chronic atopic dermatitis is exemplified by lichenified plaques and papules with excoriations. Atopic dermatitis patients are also at higher risk for skin infections, including bacterial and viral superinfections. Conventional therapy includes avoidance of irritants and potential allergens, as well as continued hydration of the skin with thick emollients. Topical corticosteroids and topical immunomodulators are often used primarily. Other therapies including phototherapy, antimicrobials, antihistamines, and systemic immunosuppressives are also options in certain situations.

A bilateral comparison study of pimecrolimus cream 1% and a topical medical device cream in the treatment of patients with atopic dermatitis.

Corticosteroids are the mainstay of therapy for atopic dermatitis, but long-term use is associated with adverse effects. We sought to evaluate the clinical efficacy of two steroid-sparing creams for atopic dermatitis. Twenty patients were enrolled in an investigator-blinded, bilateral comparison study. Patients applied pimecrolimus cream twice daily to a target lesion on one side of the body and also applied a topical medical device cream three times daily on a symmetrical target lesion on the opposite side of the body for four weeks. Clinical assessments including Physician Global Assessment (PGA), Target Lesion Symptom Score (TLSS), subject self-assessment and digital photography were performed at the baseline, 2 week, and 4 week visits. Seventy-five percent of patients (pimecrolimus, 15 of 20; topical medical device, 15 of 20) were rated "clear" (0) or "almost clear" (1) by PGA for both medications after four weeks. Percent improvement of the PGA from randomization for pimecrolimus cream and the topical medical device cream were 72.50 and 71.67 respectively (P=0.9283). PGA scores decreased significantly from baseline for both treatments (P=0.004). Overall, there was no statistically significant difference between treatment groups for PGA scores throughout the study (P=0.8236). No cutaneous side effects were noted. Our study was limited by a small sample size and lack of double-blinding; however, both treatments were found to be safe and effective in treating atopic dermatitis over four weeks. Significant improvements were noted for all efficacy variables. In conclusion, a lipid-rich, non-steroidal, topical medical device cream was as effective in improving atopic dermatitis as pimecrolimus cream.

Bilateral comparison study of pimecrolimus cream 1% and a ceramide-hyaluronic acid emollient foam in the treatment of patients with atopic dermatitis.

Topical corticosteroids have been the mainstay of treatment for atopic dermatitis (AD) over the last decade, especially in the setting of acute flares. However, heavy and prolonged use of topical corticosteroid is undesirable as it is associated with side effects such as, skin atrophy, telangiectasia, striae, steroid-induced dermatoses, rosacea, acne exacerbation, and in some severe and rare cases, systemic effects such as hypothalamic-pituitary-adrenal axis suppression, growth retardation and ocular problems. Non-steroidal ant-inflammatory agents specific for the treatment of AD (topical calcineurin inhibitors, or TCIs) are now available and they are a viable alternative to topical corticosteroids in treating dermatitis of the face, neck, eyelids, and intertriginous areas where there is a greater risk of the steroid-induced side effects. More recently, medical device emollients have entered the marketplace. These medical devices provide, but are not limited to, anti-oxidant, anti-protease, anti-inflammatory activity, and aid in restoring the natural balance of lipids, which is one of the causes of the epidermal abnormalities seen with AD. The present study evaluated the short-term effectiveness and appeal of a non-steroidal medicated device foam as compared to pimecrolimus cream 1% in the treatment of AD within a wide age group of subjects with active disease at baseline. In this study, both pimecrolimus and the medical device foam exhibited efficacy in mild-to-moderate AD. Primary efficacy was measured by IGA. After four weeks of treatment with the medical device foam, 82% of target lesions were scored "clear" (0) or "almost clear" (1) compared to 71% of target lesions under the pimecrolimus arm. This study confirmed that pimecrolimus cream 1% and the medical device foam work well in the treatment of AD in both adults and children with no associated adverse effects.

A Randomized, Double-blind, Placebo-controlled Study to Evaluate the Safety and Efficacy of Ammonium Lactate Lotion 12% and Halobetasol Propionate Ointment 0.05% in the Treatment and Maintenance of Psoriasis.

Consensus recommends a gradual reduction in the frequency or steroid potency of topical corticosteroids following clinical improvement in the treatment of psoriasis, although no established guidelines have been developed. The authors sought to evaluate a combination regimen in the treatment and maintenance of psoriasis. Patients with mild-to-moderate psoriasis were enrolled (n=55) in a randomized, double-blind, placebo-controlled study using ammonium lactate lotion and halobetasol ointment. Those with initial improvement of target plaques after two weeks of combination treatment twice daily were randomized to a maintenance phase (n=41). Patients applied ammonium lactate lotion twice daily everyday and either placebo ointment (n=20) or steroid ointment (n=21) twice daily on weekends only. Forty-one of 55 patients (74.6%) were rated as "clear" (0) or "almost clear" (1) after two weeks of combination treatment. In the maintenance phase, the probability of physician global assessment worsening at six weeks in the steroid group was only 10 percent while in the placebo group the probability rose to 75 percent (p<0.0001). The probability of physician global assessment worsening climbed to 100 percent by 14 weeks in the placebo group while only increasing to 29 percent in the steroid group (p<0.0001). Twelve patients at study termination still had not worsened. Worsening of the physician global assessment index was more likely (HR 7.8 [2.84, 21.43]) in the placebo group than in the steroid group (p<0.0001). No cutaneous side effects, such as steroid atrophy or irritation, were noted. Combination treatment effectively cleared plaque psoriasis initially, and ammonium lactate twice daily everyday with weekend-only applications of halobetasol ointment effectively sustained the initial improvement for a significantly longer period of time when compared with placebo without demonstrating any significant side effects, such as steroid atrophy.

COX-2 inhibition and inhibition of cytosolic phospholipase A2 increase CD36 expression and foam cell formation in THP-1 cells.

Cardiovascular safety of cyclooxygenase (COX)-2-selective inhibitors and nonselective nonsteroidal anti-inflammatory drugs (NSAIDs) is of worldwide concern. COX-2 inhibitors and NSAIDs act by inhibiting arachidonic acid metabolism to prostaglandins. They confer a cardiovascular hazard manifested as an elevated risk of myocardial infarction. Mechanisms underlying these cardiovascular effects are uncertain. Here we determine whether interference with cytosolic phospholipase A2 (cPLA-2) or COX-2 through pharmacologic blockade or silencing RNA impacts expression of scavenger receptor CD36 and scavenger receptor A, both involved in cholesterol uptake in monocytes and macrophages. THP-1 human monocytes and human peripheral blood mononuclear cells were exposed to celecoxib, a COX-2 selective inhibitor currently in clinical use, and to arachidonyl trifluoromethyl ketone (AACOCF3), an arachidonic acid analog that selectively inhibits cPLA-2. Celecoxib and AACOCF3 each upregulated expression of CD36, but not scavenger receptor A, as determined by quantitative PCR and immunoblotting. Silencing of cPLA-2 or COX-2 had comparable effects to pharmacologic treatments. Oil red O staining revealed a profound increase in foam cell transformation of THP-1 macrophages exposed to either celecoxib or AACOCF3 (both 25 µM), supporting a role for the COX pathway in maintaining macrophage cholesterol homeostasis. Demonstration of disrupted cholesterol balance by AACOCF3 and celecoxib provides further evidence of the possible mechanism by which COX inhibition may promote lipid overload leading to atheromatous lesion formation and increased cardiovascular events.