Localizing From Classification: Self-Directed Weakly Supervised Object Localization for Remote Sensing Images.

In recent years, object localization and detection methods in remote sensing images (RSIs) have received increasing attention due to their broad applications. However, most previous fully supervised methods require a large number of time-consuming and labor-intensive instance-level annotations. Compared with those fully supervised methods, weakly supervised object localization (WSOL) aims to recognize object instances using only image-level labels, which greatly saves the labeling costs of RSIs. In this article, we propose a self-directed weakly supervised strategy (SD-WSS) to perform WSOL in RSIs. To specify, we fully exploit and enhance the spatial feature extraction capability of the RSIs' classification model to accurately localize the objects of interest. To alleviate the serious discriminative region problem exhibited by previous WSOL methods, the spatial location information implicit in the classification model is carefully extracted by GradCAM ++ to guide the learning procedure. Furthermore, to eliminate the interference from complex backgrounds of RSIs, we design a novel self-directed loss to make the model optimize itself and explicitly tell it where to look. Finally, we review and annotate the existing remote sensing scene classification dataset and create two new WSOL benchmarks in RSIs, named C45V2 and PN2. We conduct extensive experiments to evaluate the proposed method and six mainstream WSOL methods with three backbones on C45V2 and PN2. The results demonstrate that our proposed method achieves better performance when compared with state-of-the-arts.

Achieving Better Category Separability for Hyperspectral Image Classification: A Spatial-Spectral Approach.

The task of hyperspectral image (HSI) classification has attracted extensive attention. The rich spectral information in HSIs not only provides more detailed information but also brings a lot of redundant information. Redundant information makes spectral curves of different categories have similar trends, which leads to poor category separability. In this article, we achieve better category separability from the perspective of increasing the difference between categories and reducing the variation within category, thus improving the classification accuracy. Specifically, we propose the template spectrum-based processing module from spectral perspective, which can effectively expose the unique characteristics of different categories and reduce the difficulty of model mining key features. Second, we design an adaptive dual attention network from spatial perspective, where the target pixel can adaptively aggregate high-level features by evaluating the confidence of effective information in different receptive fields. Compared with the single adjacency scheme, the adaptive dual attention mechanism makes the ability of target pixel to combine spatial information to reduce variation more stable. Finally, we designed a dispersion loss from the classifier's perspective. By supervising the learnable parameters of the final classification layer, the loss makes the category standard eigenvectors learned by the model more dispersed, which improves the category separability and reduces the rate of misclassification. Experiments on three common datasets show that our proposed method is superior to the comparison method.

Surgical Treatment of Wilkie's Syndrome by Vascular Transposition.

Introduction Superior mesenteric artery syndrome (SMAS), also called mesenteric duodenal compression syndrome, Wilkie's syndrome, chronic duodenal ileus or cast syndrome, is a rare clinical condition defined as a compression of the third portion of the duodenum in between the SMA and abdominal aorta (AA), due to narrowing of the space between them. SMAS is primarily attributed to loss of the intervening mesenteric fat pad, leading to partial or complete duodenal obstruction. Its manifestations are complex and non-specific, including postprandial epigastric pain, nausea, vomiting, early satiety, weight loss and anorexia. SMAS may present as an acute syndrome, or it may have an insidious onset with chronic symptoms. SMAS mainly affects females between 10 and 40 years of age. This study aims to discuss the safety and efficacy of vascular decompression of the duodenum by infrarenal transposition of SMA. Methods This single-centre prospective clinical study analysed 37 patients with Wilkie's syndrome who underwent infrarenal transposition of the SMA between January 2012 and December 2021. The indications for the surgery were severe weight loss, uncontrolled upper abdominal pain, vomiting and other gastrointestinal (GI) symptoms that were severely debilitating to patients' daily lives, along with radiological findings such as aortomesenteric angle < 25°, aortomesenteric distance <8 mm and distention of proximal part of the duodenum and the stomach. Ten patients (27%) concurrently had Nutcracker syndrome and seven patients (18.9%) had Dunbar syndrome (median arcuate ligament syndrome). Three female patients (8.1%) had all three above-mentioned vascular compression syndromes, which were treated in the same surgery. One male patient (2.7%) was after a laparoscopic duodenojejunostomy with symptoms that relapsed three months postoperatively, which was cured after the infrarenal transposition of SMA. Results Technical operative and clinical success were achieved in all patients. There were no cases of anastomotic failure, SMA thrombosis or intestinal ischemia. All of the patients are currently living symptom-free. One patient (2.7%), four days postoperatively, had a lymphocele formed in the retroperitoneum, which was successfully drained by a CT-guided percutaneous pigtail catheter. Another patient (2.7%) after three months of surgery needed a re-laparotomy for adhesive obstruction of the second part of the duodenum and was treated by adhesiolysis and omentoplasty. One patient (2.7%), 2-year postoperatively, had a proximal SMA stenosis up to 60% where drug-eluting balloon percutaneous transluminal angioplasty (DEB PTA) was performed successfully. Finally, the upper GI symptoms were resolved in all 37 patients (100%). Conclusion Wilkie's syndrome, although rare, is frequently late-diagnosed or underdiagnosed. In cases of failure of conservative therapy, infrarenal transposition of the SMA can be considered a safe and feasible surgical option with more physiologically favourable outcomes comparable to gastrointestinal bypasses, especially in patients concurrently suffering from Nutcracker syndrome. Simultaneously, it also restores physiologic duodenal passage of gastroduodenal content without the need of creating a digestive tract anastomosis. To our best knowledge, we have the highest number of SMA transposition surgeries performed in a single centre for the treatment of Wilkie's syndrome.

KL-MOB: automated COVID-19 recognition using a novel approach based on image enhancement and a modified MobileNet CNN.

The emergence of the novel coronavirus pneumonia (COVID-19) pandemic at the end of 2019 led to worldwide chaos. However, the world breathed a sigh of relief when a few countries announced the development of a vaccine and gradually began to distribute it. Nevertheless, the emergence of another wave of this pandemic returned us to the starting point. At present, early detection of infected people is the paramount concern of both specialists and health researchers. This paper proposes a method to detect infected patients through chest x-ray images by using the large dataset available online for COVID-19 (COVIDx), which consists of 2128 X-ray images of COVID-19 cases, 8,066 normal cases, and 5,575 cases of pneumonia. A hybrid algorithm is applied to improve image quality before undertaking neural network training. This algorithm combines two different noise-reduction filters in the image, followed by a contrast enhancement algorithm. To detect COVID-19, we propose a novel convolution neural network (CNN) architecture called KL-MOB (COVID-19 detection network based on the MobileNet structure). The performance of KL-MOB is boosted by adding the Kullback-Leibler (KL) divergence loss function when trained from scratch. The KL divergence loss function is adopted for content-based image retrieval and fine-grained classification to improve the quality of image representation. The results are impressive: the overall benchmark accuracy, sensitivity, specificity, and precision are 98.7%, 98.32%, 98.82% and 98.37%, respectively. These promising results should help other researchers develop innovative methods to aid specialists. The tremendous potential of the method proposed herein can also be used to detect COVID-19 quickly and safely in patients throughout the world.

Transfer Learning to Detect COVID-19 Automatically from X-Ray Images Using Convolutional Neural Networks.

The novel coronavirus disease 2019 (COVID-19) is a contagious disease that has caused thousands of deaths and infected millions worldwide. Thus, various technologies that allow for the fast detection of COVID-19 infections with high accuracy can offer healthcare professionals much-needed help. This study is aimed at evaluating the effectiveness of the state-of-the-art pretrained Convolutional Neural Networks (CNNs) on the automatic diagnosis of COVID-19 from chest X-rays (CXRs). The dataset used in the experiments consists of 1200 CXR images from individuals with COVID-19, 1345 CXR images from individuals with viral pneumonia, and 1341 CXR images from healthy individuals. In this paper, the effectiveness of artificial intelligence (AI) in the rapid and precise identification of COVID-19 from CXR images has been explored based on different pretrained deep learning algorithms and fine-tuned to maximise detection accuracy to identify the best algorithms. The results showed that deep learning with X-ray imaging is useful in collecting critical biological markers associated with COVID-19 infections. VGG16 and MobileNet obtained the highest accuracy of 98.28%. However, VGG16 outperformed all other models in COVID-19 detection with an accuracy, F1 score, precision, specificity, and sensitivity of 98.72%, 97.59%, 96.43%, 98.70%, and 98.78%, respectively. The outstanding performance of these pretrained models can significantly improve the speed and accuracy of COVID-19 diagnosis. However, a larger dataset of COVID-19 X-ray images is required for a more accurate and reliable identification of COVID-19 infections when using deep transfer learning. This would be extremely beneficial in this pandemic when the disease burden and the need for preventive measures are in conflict with the currently available resources.

Dielectrophoresis of Amyloid-Beta Proteins as a Microfluidic Template for Alzheimer's Research.

We employed dielectrophoresis to a yeast cell suspension containing amyloid-beta proteins (Aß) in a microfluidic environment. The Aß was separated from the cells and characterized using the gradual dissolution of Aß as a function of the applied dielectrophoretic parameters. We established the gradual dissolution of Aß under specific dielectrophoretic parameters. Further, Aß in the fibril form at the tip of the electrode dissolved at high frequency. This was perhaps due to the conductivity of the suspending medium changing according to the frequency, which resulted in a higher temperature at the tips of the electrodes, and consequently in the breakdown of the hydrogen bonds. However, those shaped as spheroidal monomers experienced a delay in the Aß fibril transformation process. Yeast cells exposed to relatively low temperatures at the base of the electrode did not experience a positive or negative change in viability. The DEP microfluidic platform incorporating the integrated microtip electrode array was able to selectively manipulate the yeast cells and dissolve the Aß to a controlled extent. We demonstrate suitable dielectrophoretic parameters to induce such manipulation, which is highly relevant for Aß-related colloidal microfluidic research and could be applied to Alzheimer's research in the future.

Self-Reported Outcomes in Individuals Aged 65 and Older Admitted for Treatment to an Acute Care Surgical Service: A 6-Month Prospective Cohort Study.

OBJECTIVES: To examine health-related quality of life (HRQL) and cognitive and functional status before and after emergency surgical care in elderly adults. DESIGN: Six-month prospective cohort study. SETTING: Acute care and emergency surgery service at a single, academic tertiary care center, Edmonton, Alberta, Canada. PARTICIPANTS: Admitted individuals aged 65 and older (mean age 77.8 ± 7.9, 52% female) or their surrogates. MEASUREMENTS: Abbreviated Mental Test Score-4 (AMTS), Barthel Index, Vulnerable Elders Survey (VES-13), and EuroQol-5 Dimensional Scale (EQ-5D) completed by participants or their surrogates within 24 hours of admission to the hospital and 6 months after discharge. Paired t-tests and McNemar tests were used to assess the difference between baseline and 6 months. RESULTS: One hundred fifty-five consecutive individuals (including 16 surrogates) were enrolled. Sixteen (10%) died within 6 months of discharge, and 116 (75%, including 18 surrogates) completed a follow-up assessment 6 months after discharge. Cognitive status improved substantially over 6 months, with 72 (52%) of participants having AMTS scores showing cognitive impairment at baseline and four (4%) having AMTS scores showing cognitive impairment at 6 months (P < .001). There was no statistically significant change from baseline on the Barthel Index, VES-13, or EQ-5D. CONCLUSION: There was significant cognitive improvement in older adults after surgical hospitalization. HRQL improved back to age-matched population norms. These results suggest that elderly adults admitted for emergency surgery have good cognitive, functional, and HRQL outcomes.

Eyelid skin disinfecting and conjunctival bacteria in cataract surgery.

PURPOSE: To evaluate the effect on conjunctival micro-organisms of chlorhexidine alcohol 0.5% eyelid wash in cataract patients. METHODS: In this single-centre experimental study, 120 patients scheduled to undergo cataract operation were cultured in the conjunctiva before the operation and after the insertion of the intraocular lens. Bacterial samples for aerobic and anaerobic cultures were taken with a polyvinyl alcohol pad and transported in peptone water to the microbiological laboratory where they were analysed in a blinded fashion. Half of the patients received a standardized eyelid scrub with chlorhexidine alcohol 0.5%, while the rest did not. The conjunctiva of all patients was rinsed preoperatively with chlorhexidine solution 0.05%. RESULTS: In all the between-group comparisons, the treated eyelid group had slightly lower bacterial counts than the control group on the postoperative samples, but differences were far from statistically significant. Within-group comparisons revealed statistically significant reduction in colonies and species between the preoperative and postoperative sampling. CONCLUSION: In this experimental culture study, no evidence was found to suggest that preoperative eyelid chlorhexidine alcohol wash is effective in further reducing bacterial counts in the conjunctiva when the conjunctiva itself is rinsed preoperatively with chlorhexidine solution. The present results, however, do not contradict that the wash may be of some marginal help in reducing bacterial contamination of the operation field. Consequently, this part of the prophylactic protocol is still in practice in our institution.

Early Outcomes of Roux-en-Y Gastric Bypass in a Publically Funded Obesity Program.

Background. There is limited literature assessing the outcomes of bariatric surgery in a publically funded, North American, multidisciplinary bariatric program. Our objective was to assess outcomes of roux-en-Y gastric bypass (RYGB) in a publically funded bariatric program through a retrospective review of patient records. Methods. 293 patients spent a median of 13 months attending a multidisciplinary obesity clinic prior to undergoing laparoscopic RYGB surgery. The hospital was a Canadian, publically funded, level 2 trauma center with university teaching services. Results. 79% of the patients were female and the average BMI at first visit to clinic was 55.3 kg/m2. The average decrease in BMI was 19.2 ± 0.9 kg/m(2). This was an average absolute weight loss of 56.1 kg or 35.5% of initial weight. The average excess weight loss was 63.4 ± 20.4%. Improvement or resolution of obesity related comorbidities occurred in 65.9% of type 2 diabetics and in 50% of hypertensive patients. Conclusion. Despite this being an unconventional setting of a publically funded program in a large Canadian teaching hospital, early outcomes following RYGB were appropriate in severely obese patients. Ongoing work will identify areas of improvement for enhanced efficiencies within this system.