Advancing breast ultrasound diagnostics through hybrid deep learning models.

Today, doctors rely heavily on medical imaging to identify abnormalities. Proper classification of these abnormalities enables them to take informed actions, leading to early diagnosis and treatment. This paper introduces the "EfficientKNN" model, a novel hybrid deep learning approach that combines the advanced feature extraction capabilities of EfficientNetB3 with the simplicity and effectiveness of the k-Nearest Neighbors (k-NN) algorithm. Initially, EfficientNetB3, pre-trained on ImageNet, is repurposed to serve as a feature extractor. Subsequently, a GlobalAveragePooling2D layer is applied, followed by an optional Principal Component Analysis (PCA) to reduce dimensionality while preserving critical information. PCA is used selectively when deemed necessary. The extracted features are then classified using an optimized k-NN algorithm, fine-tuned through meticulous cross-validation.Our model underwent rigorous training using a curated dataset containing benign, malignant, and normal medical images. Data augmentation techniques, including rotations, shifts, flips, and zooms, were employed to help the model generalize and efficiently handle new, unseen data. To enhance the model's ability to identify the important features necessary for accurate predictions, the dataset was refined using segmentation and overlay techniques. The training utilized an ensemble of optimization algorithms-SGD, Adam, and RMSprop-with hyperparameters set at a learning rate of 0.00045, a batch size of 32, and up to 120 epochs, facilitated by early stopping to prevent overfitting.The results demonstrate that the EfficientKNN model outperforms traditional models such as VGG16, AlexNet, and VGG19 in terms of accuracy, precision, and F1-score. Additionally, the model showed better results compared to EfficientNetB3 alone. Achieving a 100 % accuracy rate on multiple tests, the EfficientKNN model has significant potential for real-world diagnostic applications. This study highlights the model's scalability, efficient use of cloud storage, and real-time prediction capabilities, all while minimizing computational demands.By integrating the strengths of EfficientNetB3's deep learning architecture with the interpretability of k-NN, EfficientKNN presents a significant advancement in medical image classification, promising improved diagnostic accuracy and clinical applicability.

Transport of Mobile Particles in Heavy Metal Contaminated Soil with Simulated Acid Rain Leaching.

The soil contaminated with heavy metals requires special attention due to its adverse effects on health of human and animals. The effects of simulated acid rain with different pH values on transport of heavy metal in contaminated soil of Phyllostachys pubescens forest were studied by indoor leaching column test. The results revealed that particle size of soil was mainly concentrated in range of more than 50 µm. The content of heavy metals in particles less than 50 µm was relatively high. The Pb and Zn were mainly adsorbed on colloidal particles and were transported during simulated acid rain. The release of Fe and Al increased the release of particulate matter in soil leaching solution. The mobility of Zn was increased at low pH.

Mitigation of Cd accumulation in rice with water management and calcium-magnesium phosphate fertilizer in field environment.

Pollution of Cd has seriously threatened environmental safety and human health. The field experiment was conducted to investigate the effects of calcium-magnesium phosphate fertilizer and water management on bioavailability of Cd in soils and its accumulation in rice. The results revealed that continuous flooding has enhanced soil pH from 5.10 to 5.72 and reduced soil redox potential (Eh) from 164 to - 60 mV. Application of calcium-magnesium phosphate fertilizer has significantly raised soil pH from 5.10 to 6.45 (P < 0.05). The treatment of calcium-magnesium phosphate fertilizer and continuous flooding has reduced available content of Cd in soils by 28.57%. The content of Cd in brown rice was significantly diminished by 51.36% (P < 0.05). The continuous flooding has promoted formation of residual Cd in soil with application of calcium-magnesium phosphate fertilizer. The biomass and grain production of rice was not significantly decreased compared with control.

Incidence of blinding vitamin A deficiency in North West Frontier Province and its adjoining Federally Administered Tribal Areas, Pakistan.

PURPOSE: To identify children aged 0-72 months with blinding xerophthalmia in the North West Frontier Province (NWFP) and its adjoining Federally Administered Tribal Areas (FATA) in Pakistan, using a new surveillance system, and to describe socio-economic and other characteristics of reported cases. METHODS: All ophthalmologists in district headquarter hospitals in NWFP and agency headquarter hospitals in FATA were requested to participate in a surveillance programme for the detection of blinding xerophthalmia (corneal xerosis, corneal ulcers, keratomalacia and corneal scars related to vitamin A deficiency) for children aged 0-72 months. The surveillance included reporting of each case of blinding xerophthalmia presenting to the participating eye units over a period of 12 months and recording their socio-economic and other related characteristics. RESULTS: Seventy six cases of blinding xerophthalmia were reported and there was almost no gender difference. Children came from 19 of 31 districts and agencies in NWFP and FATA with a working ophthalmologist. The commonest age at presentation was 25-48 months and almost half of them belonged to households with a monthly income of < Pak Rs. 1200 (US $ < 20). Precipitating systemic illnesses were diarrhea and fever with cough in 63.2% and 56.6% cases, respectively. Almost a third of children aged > 9 months had not been immunized against measles. A quarter of all children had not been adequately breast-fed, and 7 children (9%) had never been breastfed. Majority of the cases presented during summer and autumn months. CONCLUSIONS: Our surveillance data provide evidence that blinding xerophthalmia does occur in children in NWFP and its adjoining FATA. The surveillance system identified high risk age groups, systemic illnesses, seasons and community groups. The detection system detects minimal numbers of cases and it is likely that the problem is even more severe.

Lens opacities in adults in Pakistan: prevalence and risk factors.

PURPOSE: To investigate the prevalence and risk factors for lens opacity (LO) amongst a nationally representative sample of the adult population of Pakistan. METHODS: This national study of blindness and visual impairment (adults > or =30 years) used multistage, stratified, cluster random sampling. Grading of LO was conducted using the Mehra/Minassian classification system. LO, partly or wholly obscuring the red reflex, or previous cataract surgery were indicators of opacity. RESULTS: 16,402 (94.7%) adults were included in this analysis (study conducted 2002-2003). A total of 4,096 (standardized prevalence 20.9%, 95%CI: 20.3, 21.5%) adults were found to have LO. The highest prevalence of LO was found in Punjab province (22.2%), the lowest in Balochistan Province (18.0%). Significant positive associations were increasing age (multivariable odds ratio (OR) 3.2: 95%CI: 3.1, 3.4), hypertension (OR 1.2, 95%CI: 1.1, 1.3), history of diabetes (OR 2.6: 95%CI 2.0, 3.2) and smoking (OR 1.3: 95%CI: 1.1, 1.5). Higher body mass index (BMI) (OR 0.8: 95%CI 0.7, 0.9, heavy vs. normal BMI) and attendance to school (OR 0.6: 95%CI: 0.5, 0.8) were associated with lower risk of LO. Individuals in districts classified as hot were at significantly increased risk (OR 1.3: 95%CI: 1.1, 1.5), and those in wet districts (> 600 mm annual rainfall) had lower odds than individuals living in dry districts (OR 0.7: 95%CI: 0.6, 0.8). CONCLUSIONS: Almost a fifth of the adult population had LO. Significant positive associations were age, smoking status, hypertension, diabetes, and increased deprivation level. Protective factors included high BMI and educational achievement. The climatic associations offer novel hypothesis for further research into cataractogenesis.

Perceptions of eye health in schools in Pakistan.

BACKGROUND: Research exploring children's and their teachers' perceptions of eye health is lacking. This paper reports for the first time on perceptions of primary schoolchildren and their teachers of healthy and diseased eyes, things that keep eyes healthy and damage them, and what actions to be taken in case of an eye injury. METHODS: Using draw and write technique, 160 boys and girls (9-12 years old) attending four primary schools in Abbottabad district, northern Pakistan, were invited to draw pictures in response to a set of semi-structured questions and then label them. Sixteen teachers who were currently teaching the selected students were interviewed one-on-one. RESULTS: Analysis of text accompanying 800 drawings and of the interview scripts revealed that most children and teachers perceived healthy eyes to be those which could see well, and diseased eyes to be those which have redness, watering, dirty discharge, pain, and itching; or those which have "weak eyesight" and blindness. Among things that students and teachers thought damage the eyes included sun, television, and sharp pointed objects, particularly pencils. Teachers noted that children with eye problems "have difficulty seeing the blackboard well", "screw up their eyes", and "hold their books too close". CONCLUSION: We conclude that schoolchildren and their teachers had a good knowledge of eye health, but many of them had serious misconceptions e.g., use of kohl, medicines and eye drops keeps eyes healthy. Kohl is an important source of lead and can reduce children's intelligence even at low blood levels. Health education in schools must take into account children's existing knowledge of and misconceptions about various aspects of eye health. Such steps if taken could improve the relevance of eye health education to schoolchildren.

Prevalence and causes of blindness and low vision in a rural setting in Pakistan.

PURPOSE: To determine the prevalence and causes of blindness and visual impairment in people 40 years of age and older in Budni, Peshawar, Pakistan. METHODS: A population-based cross-sectional study was carried out involving 1,106 men and women 40 years of age and older in a rural area in Pakistan's North West Frontier Province (NWFP). All subjects with a presenting visual acuity < 6/18 in either eye were referred to a centralized clinic for a standardized eye examination that included refraction and dilated fundal examination. The main outcome was blindness (presenting visual acuity < 3/60 in the better eye) and low vision (presenting VA < 6/18-3/60 in the better eye). RESULTS: Of 1,106 people examined, 21 (1.9%; 95% CI: 1.1-2.7%) were blind, while another 27 (2.4%) and 62 (5.5%) subjects had severe visual impairment (< 6/60-3/60) and visual impairment (< 6/18-6/60), respectively. Women, as compared to men, had a higher prevalence of visual impairment and severe visual impairment; but they had a lower prevalence of blindness (1.6 vs. 2.2%); however, the difference was not statistically significant (0.6%; 95% CI: -0.9-2.1%). Similarly farmers had the highest prevalence of blindness. The leading cause of blindness and low vision was cataract, which accounted for 14 of 21 (66.6%) cases of blindness and 49 of 89 (55.5%) cases of low vision. The second leading cause of blindness was uncorrected aphakia. CONCLUSION: Much of the blindness was due to unoperated cataract and uncorrected aphakia. Thus, there is an urgent need to develop ways in which cataract surgical output could be increased, and glasses provided to those who need them.

Classification of 154 clinical cases of vitamin A deficiency in children (0-15 years) in a tertiary hospital in North West Frontier Province Pakistan.

OBJECTIVE: To share our experience of 154 cases of clinical vitamin A deficiency in Pakistani children. METHODS AND RESULTS: The data on age, sex, date of presentation and clinical stage of vitamin A deficiency was recorded. A total of 154 children aged 0-15 years suffering from clinical vitamin A deficiency (xerophthalmia) were recorded. One hundred and twenty three (79.9%) children were 0-6 years with 99 (64.3%) male children. Blinding xerophthalmia (corneal xerosis, corneal ulcers ad keratomalacia) was present in 94 (61%) children. Thirty nine percent children presented in summer, 28% in autumn, 20% in winter and 13% in spring respectively. CONCLUSION: This report indicates that clinical cases of vitamin A deficiency in children do occur in Pakistan. Children under 6 years of age are the most vulnerable age group and there is a seasonal variation in presentation. Both life and sight of these children are at risk. Community based studies may be helpful to identify the magnitude of the problem and possible risk factors at national, provincial and district levels.