Comparative analysis of features and classification techniques in breast cancer detection for Biglycan biomarker images.

BACKGROUND: Breast cancer (BC) is considered the world's most prevalent cancer. Early diagnosis of BC enables patients to receive better care and treatment, hence lowering patient mortality rates. Breast lesion identification and classification are challenging even for experienced radiologists due to the complexity of breast tissue and variations in lesion presentations. OBJECTIVE: This work aims to investigate appropriate features and classification techniques for accurate breast cancer detection in 336 Biglycan biomarker images. METHODS: The Biglycan biomarker images were retrieved from the Mendeley Data website (Repository name: Biglycan breast cancer dataset). Five features were extracted and compared based on shape characteristics (i.e., Harris Points and Minimum Eigenvalue (MinEigen) Points), frequency domain characteristics (i.e., The Two-dimensional Fourier Transform and the Wavelet Transform), and statistical characteristics (i.e., histogram). Six different commonly used classification algorithms were used; i.e., K-nearest neighbours (k-NN), Naïve Bayes (NB), Pseudo-Linear Discriminate Analysis (pl-DA), Support Vector Machine (SVM), Decision Tree (DT), and Random Forest (RF). RESULTS: The histogram of greyscale images showed the best performance for the k-NN (97.6%), SVM (95.8%), and RF (95.3%) classifiers. Additionally, among the five features, the greyscale histogram feature achieved the best accuracy in all classifiers with a maximum accuracy of 97.6%, while the wavelet feature provided a promising accuracy in most classifiers (up to 94.6%). CONCLUSION: Machine learning demonstrates high accuracy in estimating cancer and such technology can assist doctors in the analysis of routine medical images and biopsy samples to improve early diagnosis and risk stratification.

The association between mobile phone use and neuromusculoskeletal complaints.

BACKGROUND: Mobile phones have become an essential component in our life. There are many concerns about the effects of prolonged mobile phone use on the upper extremity. OBJECTIVE: This research aims to investigate the effects of prolonged mobile phone use on the neuromusculoskeletal system of the upper extremity in young adults in Jordan. METHODS: A total of 228 subjects (age = 20.7±1.27 years) participated in this study. A structured questionnaire was used to collect data on sociodemographic variables, mobile phone size, hand/finger used for typing, duration of use daily, and the presence of musculoskeletal pain/syndromes and their location. RESULTS: Statistically significant relations (p < 0.05) were found between the prolonged use of mobile phones and self-reported pain in neck/upper back, hands, and symptoms of cubital syndrome; between smaller hand-to-phone ratio and pain in neck/upper back and shoulders with the higher frequency of pain in females. Females and subjects with a smaller hand-to-phone ratio tend to use two hands (p < 0.05). CONCLUSION: The results show an association between mobile phone use and self-reported pain. This study provides guidance for mobile phone manufacturers and regulatory agencies on the potential effects, the need for awareness programs that alert to musculoskeletal complaints, choosing the appropriate mobile phone, and changing using habits. Further investigations on a larger sample with quantitative measures on the kinematics and muscular activities are necessary to generalize and better interpret the results.

Validation of earlobe site as an alternative blood glucose testing approach.

BACKGROUND: Drawing blood from the fingertips for glucose testing is painful and likely to cause tissue damage over time. Earlobes are an alternative site for glucose measurement. OBJECTIVE: This work aims to validate the earlobe as an alternate test site for blood glucose testing by demonstrating valid and reliable statistically significant differences between the earlobes and standard reference sites. METHODS: Blood glucose concentrations from 50 volunteers were measured and statistically analysed from the reference sites (forearm and fingertip) and earlobe. The analysis included: 1) one-way analysis of variance (ANOVA), 2) regression analysis, 3) Bland Altman analysis, and 4) Clarke Error Grid analysis. RESULTS: The results indicated that there is no statistically significant difference between the three blood glucose-testing methods. For the forearm-earlobe and fingertip-earlobe, all measurements were grouped around the mean of 3.7 ± 1.96 SD and 2.96± 1.96 SD, respectively. Error grid analysis showed > 97% of all earlobe and references measurements fell in Zones A and B and were in the clinically acceptable level. CONCLUSIONS: The results have shown that the earlobe is a valid substitute for blood glucose measurements.

A validated combined musculotendon path and muscle-joint kinematics model for the human hand.

Neuromusculoskeletal models provide a mathematical tool for understanding and simulating human motor control and neuromechanics. In this work, we propose a combined computational model for the musculotendon paths and muscle-joint kinematics for the human hand, including all extrinsic and intrinsic muscles. This model is implemented based on the anatomical descriptions and a human hand dissection study. The model takes joint angles as input and estimates the musculotendon lengths, length change rates, and excursion moment arms. The proposed model is simulated to generate according moment arms, which are compared with cadaver measurements available from literature in terms of similarity coefficient s. For most muscles compared, high similarity with s≥0.70 for 92% of cases is achieved between the modeled and the measured moment arms. These results suggest the correctness of modeled moment arms and imply the feasibility of modeled musculotendon paths, lengths, and length change rates.

Sub-millimetre accurate human hand kinematics: from surface to skeleton.

A highly accurate human hand kinematics model and identification are proposed. The model includes the five digits and the palm arc based on mapping function between surface landmarks and estimated joint centres of rotation. Model identification was experimentally performed using a motion tracking system. The evaluation of the marker position estimation error, which is on sub-millimetre level across all digits, underlines model quality and accuracy. Noticeably, with the development of this model, we were able to improve various modelling assumptions from literature and found a basic linear relationship between surface and skeleton rotational angles.

Eye blinking-based method for detecting driver drowsiness.

Drowsy driving is a major cause of traffic accidents. Eye blinking is considered as important evidence of driver drowsiness. In this paper, a portable and low cost device for monitoring a driver's drowsiness is proposed. The proposed system consists of two main parts that detect eye blinking based on IR sensors mounted on eyewear. Depending on the reflected and absorbed IR radiation, this system detects and classifies the eye blinking into normal blinking (NB) or prolonged blinking (PB). The detected prolonged blinking is used to trigger an audio/visual alarm system which draws the driver's attention back. The system was simulated initially by LabVIEW® software. Moreover, the system was bench tested on 15 adult volunteers; eye blinking were detected and classified successfully for all subjects. The results of this research are promising and additional investigation is required to further improve the method.