Learning Strategies Research Articles

Exploring experiences of house officers in bridging gap in translation of knowledge into practice using different learning strategies in Prosthodontics: an in-depth qualitative approach

Objective: To explore the lived experiences of Prosthodontics house officers in bridging gap in translation of knowledge into practice using different learning strategies in Prosthodontics. Methodology: A qualitative study was conducted by using phenomenological approach; the individual face to face in depth interview was conducted at the workplace of the participants. Purposeful sampling was considered among the Prosthodontics House Officers who have successfully completed their clinical rotation in the department of Prosthodontics. Using NIVIVO software, a thematic analysis was done considering Braun and Clark (2006) thematic analysis. Results: Braun & Clarke steps were used for the thematic analysis. Seven themes were identified: Theme 1: Observational learning strategy; Theme 2: Case-discussion as a learning strategy; Theme 3: Feedback enhances skills; Theme 4: Laboratories procedures play vital role; Theme 5: Early clinical exposure as a learning strategy; Theme 6: Record keeping as learning strategy; Theme 7: Visual aids improve clinical skills. The participants suggested some learning strategies that can help in the development of clinical skills during clinical rotations of house job tenure in Prosthodontics in bridging the gap. Conclusion: There are learning strategies that can bridge the gap in translation of knowledge into clinical practice during the clinical rotations in Prosthodontics. These learning strategies are observation, case discussion, laboratory practice, taking feedback, early exposure to clinics and video-based learning. Key words: Theory practice gap, Prosthodontics, learning strategies, house officers, theoretical knowledge, clinical practice.

High-fidelity prediction of forming quality for self-piercing riveted joints in aluminum alloy based on machine learning

The effect of rivet length and sheet thickness on the cross-sectional formation and tensile-shear performance of self-piercing riveted joints in AA5754 aluminum alloy was examined through experimental investigation. The influence degree of joining parameters on the forming quality was analyzed. It was revealed that rivet length and sheet thickness are pivotal factors influencing the tensile-shear strength of the joint, culminating in the identification of four optimal riveting process schemes:LCh1Ah2A、LAh1Ah2B、LAh1Ah2B and LBh1Ah2B. A simulation model for self-piercing riveting was established, employing the GISSMO failure model and the modified Mohr-Coulomb (MMC) failure criteria to predict the damage and fracture of the aluminum alloy. A plethora of high-quality datasets depicting the cross-sections of the joints were derived from simulation analysis. Subsequently, the structure and hyperparameter determination method of traditional neural network prediction models were elucidated. By amalgamating the Aquila Optimization (AO) algorithm with the African Vultures Optimization Algorithm (AVOA), a hybrid optimization algorithm model known as MIC_AOAVOA was developed. This model effectively harnesses the strengths of various algorithms to augment search efficiency and optimization capabilities. Strategies for population initialization and adaptive weight adjustments were incorporated to enhance the algorithm's convergence velocity and the quality of solutions. The cauchy opposition-based learning (COBL) and fitness-distance balance (FDB) strategy further refined the composite algorithm, bolstering its global search capabilities and population diversity. Comparative analyses were performed with single algorithm models and traditional BP neural network models, with an in-depth examination of the MIC_AOAVOA_BP model's prediction outcomes. Comprehensive evaluations utilizing error statistics and composite evaluation indicators demonstrated that the model consistently achieved mean absolute percentage error (MAPE) values below 10%, correlation coefficients (R²) exceeding 0.98, and stable mean squared error (MSE) values around 0.0002 across the prediction of three metrics. These results underscore the model's high precision and stability. Consequently, the proposed enhanced model offers a solution that is more stable, accurate, and robust for the prediction of forming quality in self-piercing riveted joints within engineering applications.

Empirical analysis of language learning strategies for optimizing online language courses

Empirical analysis of language learning strategies for optimizing online language courses

Detection and identification of un-uniformed shape text from blurred video frames

<p><span lang="EN-GB">The identification and recognition of text from video frames have received a lot of attention recently, that makes many computer vision-based applications conceivable. In this study, we modify the picture mask and the original identification of the mask region convolution neural network and permit detection in three levels, including holistic, sequence, and at the level of pixels. To identify the texts and determine the text forms, semantics at the pixel and holistic levels can be used. With masking and detection, existences of the character and the word are separated and recognised. In addition, text detection using the results of 2-D feature space instance segmentation is done. Moreover, we explore text recognition using an attention-based optical character recognition (OCR) method with mask</span><span lang="EN-US"> r</span><span lang="EN-GB">egion convolution neural networks (R-CNN) to address and detect the problem of smaller and blurrier texts at the sequential level. Using attribute maps of the word occurrences in sequence to seq, the OCR method calculates the character sequence. At last, a fine-grained learning strategy is proposed to constructs models at word level using the annotated datasets, resulting in the training of a more precise and reliable model. The well-known benchmark datasets ICDAR 2013 and ICDAR 2015 are used to test our suggested methodology.</span></p>

Enhancing cognitive abilities through the POE teaching strategy in a virtual learning environment

With the introduction of information technology into education, the educational environment has undergone qualitative changes, that is, from a real learning environment to a digital learning environment. In a virtual learning environment (VLE), learners not only perceive themselves in new ways but also construct themselves in new ways. However, the POE (Prediction-Observation-Explanation) teaching strategy includes prediction, observation, and explanation. At the entry point of the learning strategy, the effective interaction between teachers and students can promote the development of cognitive ability. Therefore, this paper analyzed the perceptual entropy and matching degree through the radar target tracking algorithm, aiming at analyzing the current situation of the classroom and the advantages of POE teaching strategies. The findings of the study indicate that the POE teaching strategy greatly enhances students’ cognitive ability by teaching through prediction, observation, and interpretation. The research offers a novel contribution to the field of POE teaching strategy under a virtual cultural learning environment by expanding the comprehensive benefits of its students’ cognitive ability cultivation, exploring its impact on students’ academic performance, and offering practical recommendations for its implementation in the learning environment. The research demonstrates that POE strategies have positive effects on students’ academic performance and attitudes toward physics. Under the radar target tracking algorithm, the model matching degree and simplified entropy of the POE teaching strategy were both increasing. The mean of the model fit was 0.61, which was 0.40 higher than that of the first day, and the mean of the simplified entropy was 0.38, which was 0.12 higher than that of the first day. Finally, the cognitive ability and cognitive regulation ability of students under the POE teaching strategy were higher than those of the traditional teaching strategy. Compared with the traditional teaching strategy, cognitive ability was 9% higher, and cognitive regulation ability was 6% higher.

Development and evaluation of the active learning game "Who Am I?-Cellular Signal Transduction Edition": the positive impact on undergraduate education.

The teaching of physiology plays a crucial role in the education of health care professionals. However, traditional approaches to physiology classes in undergraduate health courses in Brazil often result in passive student participation. Research has shown that active methodologies are more effective in the learning process. In this study, we introduce the game "Who Am I?-Cellular Signal Transduction Edition" as an educational tool. This game follows a popular format with well-known rules and aims to enhance understanding of basic concepts related to hormones, cell signaling, and the functioning of the endocrine system. Our findings demonstrate that the game improves student knowledge and fosters enthusiasm and active engagement among participants. Additionally, student feedback has indicated a high level of appreciation for the game. By incorporating active learning strategies and a gamified approach, "Who Am I?-Cellular Signal Transduction Edition" provides a practical and enjoyable way of teaching physiology. This innovative educational tool has the potential to revolutionize physiology instruction. Demonstrating significant improvement in students' understanding, the game underscores its efficacy in enhancing knowledge acquisition and comprehension of cellular signaling and endocrine physiology topics.NEW & NOTEWORTHY We developed "Who Am I?-Cellular Signal Transduction Edition" to assist students in comprehending concepts of cellular signal transduction. This simple and cost-effective tool is perfect for educational settings with limited resources, and it encourages active learning for both small and large groups. Pre- and posttests have shown that it effectively enhances knowledge of hormonal actions and cellular signaling. Positive feedback from students emphasizes its value in reinforcing understanding and improving classroom engagement, making it a promising educational tool.

The Diagnostic Features of Peripheral Blood Biomarkers in Identifying Osteoarthritis Individuals: Machine Learning Strategies and Clinical Evidence.

People with osteoarthritis place a huge burden on society. Early diagnosis is essential to prevent disease progression and to select the best treatment strategy more effectively. In this study, the aim was to examine the diagnostic features and clinical value of peripheral blood biomarkers for osteoarthritis. The goal of this project was to investigate the diagnostic features of peripheral blood and immune cell infiltration in osteoarthritis (OA). Two eligible datasets (GSE63359 and GSE48556) were obtained from the GEO database to discern differentially expressed genes (DEGs). The machine learning strategy was employed to filtrate diagnostic biomarkers for OA. Additional verification was implemented by collecting clinical samples of OA. The CIBERSORT website estimated relative subsets of RNA transcripts to evaluate the immune-inflammatory states of OA. The link between specific DEGs and clinical immune-inflammatory markers was found by correlation analysis. Overall, 67 robust DEGs were identified. The nuclear receptor subfamily 2 group C member 2 (NR2C2), transcription factor 4 (TCF4), stromal antigen 1 (STAG1), and interleukin 18 receptor accessory protein (IL18RAP) were identified as effective diagnostic markers of OA in peripheral blood. All four diagnostic markers showed significant increases in expression in OA. Analysis of immune cell infiltration revealed that macrophages are involved in the occurrence of OA. Candidate diagnostic markers were correlated with clinical immune-inflammatory indicators of OA patients. We highlight that DEGs associated with immune inflammation (NR2C2, TCF4, STAG1, and IL18RAP) may be potential biomarkers for peripheral blood in OA, which are also associated with clinical immune-inflammatory indicators.

Cross-condition fault diagnosis of chillers based on an ensemble approach with adaptive weight allocation

The Heating, Ventilation and Air Conditioning (HVAC) systems are complex and prone to failures during operation, often leading to significant energy waste. Timely and accurate Fault Detection and Diagnosis (FDD) can enhance energy efficiency. The HVAC system operates under diverse conditions, data-driven models trained under existing conditions may experience performance degradation when faced with new conditions. Transfer learning offers an effective solution to this issue. This study proposes a novel transfer learning ensemble model based on adaptive weights, leveraging different transfer learning strategies to improve diagnosis performance under new conditions. Multiple cross-condition transfer learning tasks were implemented to test the proposed method, and its effectiveness was validated through multiple experiments to minimize the impact of randomness. Results showed that, compared to fine-tuning (FT), domain-adversarial neural network (DANN), and baseline models, the proposed method outperforms the other models. The average accuracy of multiple experiments improved by 0.21 % to 2.34 % compared to FT. Additionally, modifying DANN to utilize a small amount of labeled information from the target domain has led to greater overlap between the feature distributions of the source and target domains, resulting in improved performance that is close to that of FT. Finally, we analyzed the impact of target domain data volume on the performance of the four methods. The performance of the baseline model improved significantly with the increase in data volume, while the other models showed less improvement. Meanwhile, the diagnostic results of the baseline model were significantly influenced by experimental randomness when there is less training data, whereas the FT diagnostic results were relatively more stable.

SSTL-FM: Self-supervised transfer learning-based fusion model for the classification of benign-malignant lung nodules

Studies on the classification of lung nodules have been extensively conducted because of their significant role in diagnosing lung cancer. However, it is challenging to obtain several labeled nodules in the real world owing to the costly labeling. The limitations posed by a scarcity of labeled nodules in supervised learning hinder the development of more effective network models to improve classification performance. To address this challenge, we propose a novel self-supervised transfer learning strategy (SSTL) and develop a fusion model (FM) to extract the image features and clinical attributes of lung nodules by studying nodule-related data. The reconstruction task and transfer learning in the SSTL can learn robust knowledge by using non-nodules, thereby enhancing the subsequent classification task. FM effectively combines image features and clinical attributes to provide a more comprehensive representation of the nodule information, while the feature refinement module (FRM) further enhancing the fusion of these features. Furthermore, we explored the impact of varying amounts of non-nodule data on the classification performance and the ability of clinical attributes to distinguish between benign and malignant lung nodules. Extensive ablation and comparative experiments have demonstrated that the proposed method significantly improves performance by utilizing nodule-related data.

Towards Lightweight Super-Resolution With Dual Regression Learning.

Deep neural networks have exhibited remarkable performance in image super-resolution (SR) tasks by learning a mapping from low-resolution (LR) images to high-resolution (HR) images. However, the SR problem is typically an ill-posed problem and existing methods would come with several limitations. First, the possible mapping space of SR can be extremely large since there may exist many different HR images that can be super-resolved from the same LR image. As a result, it is hard to directly learn a promising SR mapping from such a large space. Second, it is often inevitable to develop very large models with extremely high computational cost to yield promising SR performance. In practice, one can use model compression techniques to obtain compact models by reducing model redundancy. Nevertheless, it is hard for existing model compression methods to accurately identify the redundant components due to the extremely large SR mapping space. To alleviate the first challenge, we propose a dual regression learning scheme to reduce the space of possible SR mappings. Specifically, in addition to the mapping from LR to HR images, we learn an additional dual regression mapping to estimate the downsampling kernel and reconstruct LR images. In this way, the dual mapping acts as a constraint to reduce the space of possible mappings. To address the second challenge, we propose a dual regression compression (DRC) method to reduce model redundancy in both layer-level and channel-level based on channel pruning. Specifically, we first develop a channel number search method that minimizes the dual regression loss to determine the redundancy of each layer. Given the searched channel numbers, we further exploit the dual regression manner to evaluate the importance of channels and prune the redundant ones. Extensive experiments show the effectiveness of our method in obtaining accurate and efficient SR models.

FedGE: Break the Scalability Limitation of Graph Neural Network With Federated Graph Embedding

Neighborhood aggregation algorithms, represented as graph convolutional networks, have attained non-negligible success in numerous topological structure-based scenarios with the assumption that the topological structure of the given graph is pre-defined and relatively small. However, a real-world graph generally is a super large graph consisting of many small graphs that are interconnected and overlapping. Which makes graph embedding in real-life industries, by nature, fall into the federated learning scheme. While current graph-based algorithms are only able to capture the individual topology of each natural graph, learning the complete structural information of the merged large graph remains challenging due to the unsustainable computational cost of graph convolutional operations. We propose a tailored federated graph embedding framework to learn the intact structural information of the numerous inherently linked small-scale graphs and the embedding of each node. We leverage graphs with around two and a half million nodes to validate the effectiveness and the correctness of the proposed framework.

Explainable federated learning scheme for secure healthcare data sharing.

Explainable federated learning scheme for secure healthcare data sharing.

Social tolerance and success-biased social learning underlie the cultural transmission of an induced extractive foraging tradition in a wild tool-using primate

The last two decades have seen great advances in the study of social learning (learning from others), in part due to efforts to identify it in the wild as the basis of behavioral traditions. Theoretical frameworks suggest that both the dynamics of social tolerance and transmission biases (or social learning strategies) influence the pathways of information diffusion in social groups. Bearded capuchins ( Sapajus libidinosus ) inhabiting the semiarid seasonal caatinga biome of the Serra da Capivara National Park (SCNP) form highly tolerant societies that possess the largest “tool-kit” described for monkeys, a feat likely facilitated by social learning. Here, we used social network analysis and an open diffusion experiment using an extractive foraging task to identify the occurrence of social learning and describe the pathways of social transmission of information in two wild primate populations. The dynamics of social tolerance outside of task introductions predicted opportunities for social learning, but it was tolerance during task introductions that predicted the actual pathways of social information diffusion. Our results also indicated that the capuchins mainly learned from others via direct observation and naïve individuals exhibited an observation bias toward successful males. This study supports the claims of cultural transmission in robust capuchins and empirically supports the role of social tolerance and social learning strategies in human and nonhuman primate cultural evolution.

Implementasi Pendidikan Agama Islam Dalam Pembiasaan Nilai-Nilai Cinta Al-Qur'an Pada Peserta Didik

Purpose- This study aims to explore how Islamic religious education is being implemented at MI Birrul Walidain Darussalam Tangerang through the values of the religion of love Al-Qur’an Learning Tahsin and Tahfidz, reveal the implications of using this strategy and reveal development plans for the future. Methods- This study uses qualitative methods to conduct field research. Primary and secondary data are two different categories of data. Techniques for collecting data include observation, interviews, and documentation. Findings - The results of this study indicate that 1). The implementation of learning to Love the Qur’an through Tahsin and Tahfidz at MI Birrul Walidain Batuceper has 3 aspects, namely in terms of planning, implementation and evaluation, 2). The implementation of the use of learning the Love of the Qur’an, namely Tahsin and Tahfidz at MI Birrul Walidain Batuceper is that the teacher can deepen the knowledge of the Qur’an and repeat the memorization that has been owned, reduces student difficulties, students have better morals, understand more content of verses, more disciplined, hardworking, patient, independent, honest, grades can be on target, achievers, have a good impact on the environment, a school atmosphere that loves the Qur’an, parents of students participate in Tahsin learning, and alumni become donors, and 3 ). The plan for developing the Tahsin and Tahfidz Al-Qur’an learning strategy at MI Birrul Walidain Batuceper for the future is to make a special Tahsin and Tahfidz program, hold seminars, memorize the Al-Qur’an community, train teachers, make it a special class, make certificates, graduation, acceleration, allocating funds, freeing tuition fees, facilitating students to enter universities, requirements for taking diplomas, and cooperation with Tahfidz institutions

Self-directed learning in Orthopaedic trainees and contextualisation of knowledge gaps, an exploratory study.

Self-Directed Learning (SDL) is a subjective concept with no universally agreed definition. The aim of this project was to obtain the perspectives of Orthopaedic trainee registrars on SDL. Seven Orthopaedic trainees in the Southwest Peninsula region were recruited in person by the primary author as an Orthopaedic trainee on the same programme as the participants. A one-hour Microsoft teams video interview was arranged at a mutually convenient time exploring several topics including learning resources, experiential learning, learning strategies, training programme requirements and assessment. The interviews were audio recorded and transcribed verbatim. Transcripts were coded using Corbin's and Strauss's published coding method and analysed to develop a model of SDL in this training context. Six learning stages were identified from the coded transcripts and these stages were like Knowles's stages of SDL. Participants felt less autonomous during the identification of knowledge gaps, goal setting and evaluation of learning stages. Participants perceived to have more control over the selection of strategies, and learning stages of SDL. The factors which influenced autonomy at various stages of SDL included supervisors, experiential learning opportunities, service provision competing with learning opportunities, training programme requirements and the power dynamic between trainees and supervisors. Self-advocacy by the trainees for experiential learning opportunities and for their progression was an additional action relevant to SDL that was evident. Online learning resources such as Orthobullets.com, YouTube videos, external revision courses, collaboration and webinars encouraged more autonomy with SDL. Although the United kingdom's Trauma and Orthopaedic curriculum is described as trainee led and SDL is encouraged, in practice Orthopaedic trainees felt limited autonomy with the various stages of SDL due to external factors including their level of experience and the training environment. Trainees' experiences around self-advocacy highlighted the importance of a collaborative and supportive learning culture emphasising dialogue, receiving high quality feedback, and openness for successful progression.

Improving Persuasive Writing Outcomes With Technology for Students in an Alternative Special Education Program

A study using a multiple-baseline across participants design was conducted in an alternative education setting to examine the functional relation between students’ use of a technology-based graphic organizer (TBGO) with embedded self-regulated learning strategies on the quantity and quality of their persuasive essay writing. Three third- to fifth-grade students with writing difficulties and complex behavioral needs receiving services in an alternative special education program participated in the study. The dependent variables included the number of words, number of transition words, and writing quality scores across (a) baseline (writing on the computer without the TBGO), (b) TBGO-use phase (writing on a computer with the TBGO), and (c) maintenance (writing on the computer without the TBGO). The visual and statistical analyses of data indicated all participants improved the quality of their writing, and two students also increased the quantity of their writing across phases. Students’ self-efficacy for self-regulated learning in writing ratings were also monitored throughout the study. Results indicated an overall increase in their perceived self-efficacy for writing.

Factors affecting emergency medical dispatchers decision making in stroke calls - a qualitative study.

Emergency Medical Communication Centers (EMCC) have a key role in the prehospital chain-of-stroke-survival by recognizing stroke patients and reducing prehospital delay. However, studies on EMCC stroke recognition report both substantial undertriage and overtriage. Since mis-triage at the EMCC challenges the whole chain-of-stroke-survival, by occupying limited resources for non-stroke patients or failing to recognize the true stroke patients, there is a need to achieve a more comprehensive understanding of the dispatchers' routines and experiences. The aim of this study was to explore factors affecting EMCC dispatcher's decision-making in stroke calls. A qualitative exploratory study, based on individual semi-structured interviews of 15 medical dispatchers from EMCC Oslo, Norway. Interviews were conducted during August and October 2022 and analyzed using the principles of thematic analysis. We identified four themes: [1] Pronounced stroke symptoms are easy to identify [2]. Non-specific neurological symptoms raise suspicion of acute stroke but are difficult to differentiate from other medical conditions [3]. Consistent use of the Criteria Based Dispatch (CBD) protocol may increase EMCC overtriage [4]. Contextual conditions at EMCC can affect dispatchers' decision-making process and the ability for experiential learning. Medical dispatchers at the EMCC perceive vague and non-specific stroke symptoms, such as dizziness, confusion or altered behaviour, challenging to differentiate from symptoms of other less time-critical medical conditions. They also perceive the current CBD protocol in use as less supportive in assessing such symptoms. High workload and strict EMCC response time interval requirements hinder the gathering of essential patient information and the ability to seek guidance in cases of doubt, potentially exacerbating both EMCC undertriage and overtriage. The absence of feedback loops and other strategies for experiential learning in the EMCC hampers the medical dispatcher's ability to evaluate their own assessments and improve dispatch accuracy.

Strategic Learning Principles Are Related to Academic Scores for Doctor of Physical Therapy Students

Abstract Introduction The Learning and Study Strategies Inventory (LASSI) is a self-assessment of strategic learning principles with scores representing areas for growth. The purpose of this study was to measure LASSI scores at 2 points in time for Doctor of Physical Therapy (DPT) students in a 2-year hybrid program to see if scores changed after a science of learning course and two quarters of the program. The authors then examined the relationship between LASSI scores and academic scores (DPT program GPA, anatomy, and physiology grades). Methods Retrospective descriptive analysis of LASSI scores with comparisons between baseline and after a science of learning course and two quarters of the program using a paired t-test; bivariate correlations examined the relationship of LASSI scores to academic scores. Results Data was collected for 259 matriculated DPT students. Three of the 10 LASSI scales had significant change after the science of learning course and two quarters of the program; however, the effect sizes were small (0.2 or less). Anxiety management (t(258) = 2.340, p = .020) scores improved, while both concentrations (t(258) = − 3.229, p = .001) and the use of academic resources (t(258) = − 1.999, p = .047) had lower scores. There were several LASSI scores with significant correlations to academic scores (ρ = .132 to .431). Discussion/Conclusion LASSI was related to academic scores, although the strength of the relationship was low to negligible. LASSI showed small changes over time. There could be various reasons for scores improving or declining, but the awareness of scores and change in scores can provide a solid foundation for individualized coaching to DPT students as they navigate the rigor of a graduate level professional program.

Instructor recommendations for student learning strategies and metacognition: An analysis of undergraduate biology syllabi

AbstractAn inequitable distribution of resources and opportunities for marginalized students (i.e., opportunity gaps) leads to challenges in identifying effective study behaviors, metacognition, and academic help‐seeking in higher education. While students benefit when these skills are taught explicitly through co‐curricular workshops and courses, these interventions often require significant time investment from faculty and students, underscoring a need for alternative interventions that provide students with access to resources related to these skills. Course syllabi are one potential resource that can address these needs, and we asked to what extent biology syllabi are used for this purpose. We collected a national sample of introductory biology syllabi and used content analysis to determine if syllabi are learner‐centered and whether they incorporate information on study behaviors, metacognition, and academic help‐seeking. We found that most syllabi are not learner‐centered, encourage ineffective study behaviors, did not include metacognition recommendations, and include incomplete academic help‐seeking recommendations. We make several recommendations on how to incorporate complete, accurate information regarding study behaviors, metacognition, and academic help‐seeking.

Cutting-Edge Deep Learning Strategies for Precise Ovarian Disease Detection: A Comprehensive Review

Accurate and timely diagnosis of ovary-related diseases is of paramount importance in the realm of medical imaging. Utilizing deep learning technology has emerged as a viable method to improve the precision and effectiveness of medical image segmentation, which directly impacts the detection of ovarian diseases. Ovarian disease detection through medical image segmentation is critical for early diagnosis and effective treatment. This study evaluates cutting-edge deep learning strategies using the MMOTU ovarian tumour ultrasound dataset, which includes OTU 2D and OTU CEUS subsets. We implemented various models, including U-Net, its variants with ResNet and DenseNet backbones, CR-Unet, and Ocys-Net, along with ensemble learning and transfer learning techniques. Results show that while the baseline U-Net is effective, advanced models, particularly CR-Unet and DenseNet, significantly improve segmentation accuracy. The best performance was achieved using an ensemble model and a fine-tuned pre-trained network, highlighting the potential of these approaches in enhancing the precision and reliability of ovarian disease detection.