Transformers for Energy Forecast.

Forecasting energy consumption models allow for improvements in building performance and reduce energy consumption. Energy efficiency has become a pressing concern in recent years due to the increasing energy demand and concerns over climate change. This paper addresses the energy consumption forecast as a crucial ingredient in the technology to optimize building system operations and identifies energy efficiency upgrades. The work proposes a modified multi-head transformer model focused on multi-variable time series through a learnable weighting feature attention matrix to combine all input variables and forecast building energy consumption properly. The proposed multivariate transformer-based model is compared with two other recurrent neural network models, showing a robust performance while exhibiting a lower mean absolute percentage error. Overall, this paper highlights the superior performance of the modified transformer-based model for the energy consumption forecast in a multivariate step, allowing it to be incorporated in future forecasting tasks, allowing for the tracing of future energy consumption scenarios according to the current building usage, playing a significant role in creating a more sustainable and energy-efficient building usage.

Sound Classification and Processing of Urban Environments: A Systematic Literature Review.

Audio recognition can be used in smart cities for security, surveillance, manufacturing, autonomous vehicles, and noise mitigation, just to name a few. However, urban sounds are everyday audio events that occur daily, presenting unstructured characteristics containing different genres of noise and sounds unrelated to the sound event under study, making it a challenging problem. Therefore, the main objective of this literature review is to summarize the most recent works on this subject to understand the current approaches and identify their limitations. Based on the reviewed articles, it can be realized that Deep Learning (DL) architectures, attention mechanisms, data augmentation techniques, and pretraining are the most crucial factors to consider while creating an efficient sound classification model. The best-found results were obtained by Mushtaq and Su, in 2020, using a DenseNet-161 with pretrained weights from ImageNet, and NA-1 and NA-2 as augmentation techniques, which were of 97.98%, 98.52%, and 99.22% for UrbanSound8K, ESC-50, and ESC-10 datasets, respectively. Nonetheless, the use of these models in real-world scenarios has not been properly addressed, so their effectiveness is still questionable in such situations.

Transformers for Urban Sound Classification-A Comprehensive Performance Evaluation.

Many relevant sound events occur in urban scenarios, and robust classification models are required to identify abnormal and relevant events correctly. These models need to identify such events within valuable time, being effective and prompt. It is also essential to determine for how much time these events prevail. This article presents an extensive analysis developed to identify the best-performing model to successfully classify a broad set of sound events occurring in urban scenarios. Analysis and modelling of Transformer models were performed using available public datasets with different sets of sound classes. The Transformer models' performance was compared to the one achieved by the baseline model and end-to-end convolutional models. Furthermore, the benefits of using pre-training from image and sound domains and data augmentation techniques were identified. Additionally, complementary methods that have been used to improve the models' performance and good practices to obtain robust sound classification models were investigated. After an extensive evaluation, it was found that the most promising results were obtained by employing a Transformer model using a novel Adam optimizer with weight decay and transfer learning from the audio domain by reusing the weights from AudioSet, which led to an accuracy score of 89.8% for the UrbanSound8K dataset, 95.8% for the ESC-50 dataset, and 99% for the ESC-10 dataset, respectively.

Urban Safety: An Image-Processing and Deep-Learning-Based Intelligent Traffic Management and Control System.

With the rapid growth and development of cities, Intelligent Traffic Management and Control (ITMC) is becoming a fundamental component to address the challenges of modern urban traffic management, where a wide range of daily problems need to be addressed in a prompt and expedited manner. Issues such as unpredictable traffic dynamics, resource constraints, and abnormal events pose difficulties to city managers. ITMC aims to increase the efficiency of traffic management by minimizing the odds of traffic problems, by providing real-time traffic state forecasts to better schedule the intersection signal controls. Reliable implementations of ITMC improve the safety of inhabitants and the quality of life, leading to economic growth. In recent years, researchers have proposed different solutions to address specific problems concerning traffic management, ranging from image-processing and deep-learning techniques to forecasting the traffic state and deriving policies to control intersection signals. This review article studies the primary public datasets helpful in developing models to address the identified problems, complemented with a deep analysis of the works related to traffic state forecast and intersection-signal-control models. Our analysis found that deep-learning-based approaches for short-term traffic state forecast and multi-intersection signal control showed reasonable results, but lacked robustness for unusual scenarios, particularly during oversaturated situations, which can be resolved by explicitly addressing these cases, potentially leading to significant improvements of the systems overall. However, there is arguably a long path until these models can be used safely and effectively in real-world scenarios.

Meniscal mineralisation in little spotted cats.

BACKGROUND: The aim of this study was to evaluate the stifle joints of little spotted cats in captivity using radiographic and CT studies. The hypothesis was that these animals would have meniscal mineralisation that could be detectable by imaging studies. Twelve intact little spotted cats (Leopardus tigrinus), 2 females and 10 males, aged from 1.5 to 11.11 years old and weighing 1.9-3.05 kg were studied. These animals, which were living in the Quinzinho de Barros Municipal Zoo, had no symptoms or known disease processes at the time of the study. The plain radiographs and computed tomography (CT) scans of both stifle joints were performed under general anaesthesia. Sequential transverse images were acquired on a spiral scanner. RESULTS: No signs of articular disease were observed in any of the animals. Radiographically, the meniscal mineralisation was detected as an oval radiopacity in the cranial compartment on the mediolateral projection, located within the area of the medial meniscus. On craniocaudal projection, the mineralisation was more difficult to visualise. In one of the animals, it was not possible to identify the meniscal mineralisation in either of the stifle joints. Using CT, meniscal mineralisation was best identified in the transverse plane images. CONCLUSIONS: Meniscal mineralisation appears to be a normal anatomic feature in little spotted cats.

Joint involvement in canine visceral leishmaniasis: Orthopedic physical examination, radiographic and computed tomographic findings.

This study aimed to evaluate the joint involvement in dogs with visceral leishmaniasis by means of orthopedic physical examination, radiographic and computed tomographic (CT) findings. Of the 46 evaluated dogs, an overall of 91.3 % presented joint (carpal, tarsal, elbows, and stifle) abnormalities, observed on physical examination, radiography and/or CT. In 67.3 % of the dogs orthopedic examination showed no abnormalities. Among the 31 dogs with normal orthopedic examination, 61.3 % showed radiographic and CT findings suggestive of osteoarthritis, 25.8 % presented normal radiographs with abnormalities evidenced only on CT, while 12.9 % presented normal radiographs and CT imaging. From the 15 dogs with abnormal orthopedic examination, 80 % had abnormal radiographic and CT findings suggestive of osteoarthritis, while 20 % presented normal radiographs with abnormalities evidenced only in their CT. Radiographic and CT findings included evident trabecular pattern, subchondral bone sclerosis, osteolysis, mixed bone lesions (osteolytic-proliferative lesions), soft tissue swelling around the joint (edema), joint space narrowing, bone proliferation, osteophytes, bone cyst and cartilage flap. Based on CT results the most affected joints, among those assessed, were the tarsal (80.9 %), followed by stifle (78.5 %), carpal (64.2 %), and elbows (54.7 %). Except by one dog that presented only one stifle joint compromised, the other three joints presented bilateral abnormalities in all dogs.

Chronic pain and gait analysis in dogs with degenerative hip joint disease treated with repeated intra-articular injections of platelet-rich plasma or allogeneic adipose-derived stem cells.

This prospective, comparative, randomized, horizontal, and double-blind clinical study investigated the clinical efficacy of leucocyte-poor platelet-rich plasma (PRP, n=8) or allogeneic adipose-derived stem cells (ADSC, n=8) in dogs with bilateral degenerative hip joint disease (DHJD). Sixteen dogs were treated with two intra-articular injections of PRP or ADSCs, within a 30-day interval. The Canine Brief Pain Inventory (CBPI), the Helsinki Chronic Pain Index (HCPI), and Visual Analogue Scales for pain (VAS-pain) and locomotion (VAS-loc) were assessed by the dog owners. Analysis-of-gait using a force plate, response to palpation (VAS-palp), and the descriptive numerical scale for pain (DNS) were measured by a veterinarian. The assessments were performed before (baseline), 30 and 60 days after the first treatment. Data were analyzed using the unpaired t test, paired Wilcoxon test, Fisher's exact test, and Mann-Whitney and Friedman tests (P<0.05). Compared with baseline HCPI, CBPI, VAS-pain, and VAS-palp scores reduced 41%, 52%, 51%, and 48% (P=0.0001-0.03) at 60 days in the ADSC group. In PRP-treated dogs, CBPI, VAS-loc, and DNS scores decreased by 43%, 43%, and 33% at 60 days, respectively (P=0.0003-0.011). Based on CBPI data, the rate of success at 60 days was 75% and 25% in the ADSC and PRP groups (P=0.13), respectively. Both therapies were apparently safe and effective to reduce chronic pain in dogs with bilateral DHJD during a 60-day period. However, a trend towards greater improvement was provided by the ADSC treatment.

Ultrasonographic, computed tomographic, and operative findings in dogs infested with giant kidney worms (Dioctophyme renale).

OBJECTIVE--To compare ultrasonographic, CT, and surgical findings in dogs infested with giant kidney worms (Dioctophyme renale). DESIGN--Case series. ANIMALS--15 crossbred dogs infected with D renale. PROCEDURES--Immediately after ultrasonography was performed with dogs in dorsal recumbency, sequential transverse images of the abdomen were acquired with a helical CT scanner. After plain CT, contrast CT was performed with a nonionic iodinated contrast agent. Subsequently, exploratory celiotomy was performed. RESULTS--In the corticomedullary area of the right kidney of 12 dogs, ultrasonography revealed several ring-like structures with an echogenic wall and anechoic central area in the transverse plane and arrayed as bands in the longitudinal plane. Similar structures were observed in the abdominal cavity of 10 dogs. In 13 dogs, CT revealed loss of corticomedullary differentiation in the right kidney, with discrete uptake of contrast material in the periphery of the kidney, and several ring-like or elongated structures with a hyperdense wall and hypodense center. In 11 dogs, the same structures were observed free in the abdominal cavity. Surgery revealed that 13 dogs had a damaged right kidney that required nephrectomy. Parasites were found free in the abdominal cavity of 7 dogs. CONCLUSIONS AND CLINICAL RELEVANCE--Ultrasonography and CT were effective imaging methods for detecting D renale in the kidney and less effective for detecting parasites in the abdominal cavity. Care should be taken to avoid erroneously interpreting normal structures as parasites, especially in the abdominal cavity.