Emotion detection from handwriting and drawing samples using an attention-based transformer model.

Emotion detection (ED) involves the identification and understanding of an individual's emotional state through various cues such as facial expressions, voice tones, physiological changes, and behavioral patterns. In this context, behavioral analysis is employed to observe actions and behaviors for emotional interpretation. This work specifically employs behavioral metrics like drawing and handwriting to determine a person's emotional state, recognizing these actions as physical functions integrating motor and cognitive processes. The study proposes an attention-based transformer model as an innovative approach to identify emotions from handwriting and drawing samples, thereby advancing the capabilities of ED into the domains of fine motor skills and artistic expression. The initial data obtained provides a set of points that correspond to the handwriting or drawing strokes. Each stroke point is subsequently delivered to the attention-based transformer model, which embeds it into a high-dimensional vector space. The model builds a prediction about the emotional state of the person who generated the sample by integrating the most important components and patterns in the input sequence using self-attentional processes. The proposed approach possesses a distinct advantage in its enhanced capacity to capture long-range correlations compared to conventional recurrent neural networks (RNN). This characteristic makes it particularly well-suited for the precise identification of emotions from samples of handwriting and drawings, signifying a notable advancement in the field of emotion detection. The proposed method produced cutting-edge outcomes of 92.64% on the benchmark dataset known as EMOTHAW (Emotion Recognition via Handwriting and Drawing).

Mobile-UI-Repair: a deep learning based UI smell detection technique for mobile user interface.

The graphical user interface (GUI) in mobile applications plays a crucial role in connecting users with mobile applications. GUIs often receive many UI design smells, bugs, or feature enhancement requests. The design smells include text overlap, component occlusion, blur screens, null values, and missing images. It also provides for the behavior of mobile applications during their usage. Manual testing of mobile applications (app as short in the rest of the document) is essential to ensuring app quality, especially for identifying usability and accessibility that may be missed during automated testing. However, it is time-consuming and inefficient due to the need for testers to perform actions repeatedly and the possibility of missing some functionalities. Although several approaches have been proposed, they require significant performance improvement. In addition, the key challenges of these approaches are incorporating the design guidelines and rules necessary to follow during app development and combine the syntactical and semantic information available on the development forums. In this study, we proposed a UI bug identification and localization approach called Mobile-UI-Repair (M-UI-R). M-UI-R is capable of recognizing graphical user interfaces (GUIs) display issues and accurately identifying the specific location of the bug within the GUI. M-UI-R is trained and tested on the history data and also validated on real-time data. The evaluation shows that the average precision is 87.7% and the average recall is 86.5% achieved in the detection of UI display issues. M-UI-R also achieved an average precision of 71.5% and an average recall of 70.7% in the localization of UI design smell. Moreover, a survey involving eight developers demonstrates that the proposed approach provides valuable support for enhancing the user interface of mobile applications. This aids developers in their efforts to fix bugs.

Clinical and Demographic Characteristics Including Comorbidities and Their Outcomes Among Patients Hospitalized With COVID-19 in Four Tertiary Care Hospitals Across Lahore.

Background The first case of infection with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) was diagnosed in Wuhan, China, in 2019. By the first half of 2020, coronavirus disease 2019 (COVID-19) turned into a global pandemic. Objectives The aim of this study is to describe the clinical and demographic characteristics including comorbidities and their outcomes among patients hospitalized with COVID-19 in four tertiary care hospitals across Lahore. This retrospective study was conducted at Fatima Memorial Hospital, Sir Ganga Ram Hospital, Lahore General Hospital, and Jinnah Hospital, all in Lahore, Pakistan, from May 1, 2020, to June 30, 2020. The sample size was 445, which was derived using the convenient sampling method. Clinical outcomes during hospitalization included the requirement of invasive positive pressure ventilation, need for renal replacement therapy (RRT), and death. Data regarding demographics, baseline comorbidities, important vital signs on reporting, and initial workup with results were also collected. Results A total of 445 patients' data were studied, of whom 291 (65.4%) were male patients and 154 (34.6%) female patients. The median age was 54 years (interquartile range [IQR]: 24). The most common comorbidities were hypertension (HTN) (195; 43.8%) followed by diabetes mellitus (DM) (168; 37.8%) and cardiovascular disease (CVD) (61; 13.7%). The median length of hospital stay was eight days (IQR: 3). Of the total patients, 137 (30.7%) were treated in intensive care unit settings, 40 (9%) received invasive mechanical ventilation, 40 (9%) patients had acute kidney injury, 38 (8.5%) received RRT, and 37 (8.3%) died. It was seen that more patients who were either diabetic or hypertensive received invasive mechanical ventilation as compared to those who did not have these comorbidities. The most common radiological finding on chest X-ray was the classical ground-glass appearance of COVID-19, which was found in 318 (71.4%) patients. Conclusions Patients with one or more underlying comorbidities had poor clinical outcomes compared to those with no comorbidities, with the most vulnerable group being patients with chronic kidney disease, DM, HTN, and CVD in descending order.

Frequency of Hypogonadism in Type 2 Diabetes Mellitus Patients with and without Coronary Artery Disease.

Introduction Hypogonadism is characterized by clinical and biochemical evidence of testosterone deficiency. Low testosterone levels have been reported in patients with type 2 diabetes mellitus (T2DM), which can predispose to coronary artery disease (CAD). It has been proposed that diabetic men with proven CAD have lower androgen levels than patients with normal coronary arteriograms. We conducted this study with the objective to determine the frequency of hypogonadism in patients with diabetes mellitus and its relationship with CAD. Materials and Methods It was a comparative cross-sectional study conducted at a tertiary care hospital. We recruited a total of 108 patients, divided into two groups, 54 patients in each arm of the study. Group A comprised patients with CAD, whereas group B consisted of diabetic patients without CAD. Hypogonadism was defined on the basis of erectile dysfunction clinically and total testosterone levels biochemically. CAD was diagnosed on the basis of findings of coronary angiography. Fasting blood samples were drawn and evaluated for fasting plasma glucose, HbA1c, fasting lipid profile, thyroid-stimulating hormone (TSH), serum prolactin, blood urea, serum creatinine, liver function tests (LFT), total testosterone, luteinizing hormone (LH), and follicle­stimulating hormone (FSH) levels. Hypogonadism among two study groups was compared using chi-square and serum testosterone level was compared using independent t-test with p < 0.05 considered as statistically significant. Results There were 108 subjects in the study with the mean age of 54.4 ± 4.29 (range: 22 to 68) years. The mean duration of T2DM was 12.6 ± 8.2 years. The mean BMI of patients with and without CAD was 25.7 ± 2.37 and 26.9 ± 4.21 kg/m2, respectively. There was no significant difference in waist circumference and obesity between both the groups (p-value > 0.05). Fasting plasma glucose and HbA1c in both groups were not significantly different. Testosterone levels and erectile dysfunction score were found lower in T2DM with CAD compared to T2DM patients without CAD, although this difference was not statistically significant (p-value: 0.051). The majority of the subjects with hypogonadism in both groups had a hypogonadotrophic hypogonadism (39/42, 92.9% versus 16/20, 80.0%). No statistically significant difference was seen in serum levels of LH and FSH between the study groups. The frequency of hypogonadism was found higher in the group with CAD (72.2%, 39/54) as compared with T2DM patients without CAD (37.03%, 20/54; p-value = 0.000).  Conclusion Testosterone deficiency is a significant problem of males with T2DM. Patients with CAD have markedly low levels of testosterone as compared with patients without any CAD.