Complex Problems Research Articles

Transfer learning-based English translation text classification in a multimedia network environment.

In recent years, with the rapid development of the Internet and multimedia technology, English translation text classification has played an important role in various industries. However, English translation remains a complex and difficult problem. Seeking an efficient and accurate English translation method has become an urgent problem to be solved. The study first elucidated the possibility of the development of transfer learning technology in multimedia environments, which was recognized. Then, previous research on this issue, as well as the Bidirectional Encoder Representations from Transformers (BERT) model, the attention mechanism and bidirectional long short-term memory (Att-BILSTM) model, and the transfer learning based cross domain model (TLCM) and their theoretical foundations, were comprehensively explained. Through the application of transfer learning in multimedia network technology, we deconstructed and integrated these methods. A new text classification technology fusion model, the BATCL transfer learning model, has been established. We analyzed its requirements and label classification methods, proposed a data preprocessing method, and completed experiments to analyze different influencing factors. The research results indicate that the classification system obtained from the study has a similar trend to the BERT model at the macro level, and the classification method proposed in this study can surpass the BERT model by up to 28%. The classification accuracy of the Att-BILSTM model improves over time, but it does not exceed the classification accuracy of the method proposed in this study. This study not only helps to improve the accuracy of English translation, but also enhances the efficiency of machine learning algorithms, providing a new approach for solving English translation problems.

Simultaneous well spacing and completion optimization using an automated machine learning approach. A case study of the Marcellus Shale reservoir, northeastern United States

Optimizing unconventional field development requires simultaneous optimization of well spacing and completion design. However, the conventional practice of using cross plots and sensitivity analysis via Monte Carlo simulations for independent optimization of well spacing and completion design has proved inadequate for unconventional reservoirs. This is due to the inability of cross plots to capture non-linear cross-correlations between parameters affecting hydrocarbon production, and the computational expense and difficulty of Monte Carlo simulations. Recently, automated machine learning (AutoML) workflows have been used to tackle complex problems. However, applying AutoML workflows to engineering problems presents unique challenges, as achieving high accuracy in forecasting the physics of the problem is crucial. To address this issue, a new physics-informed AutoML workflow based on the TPOT open-source tool developed that guarantees the physical plausibility of the optimum model while minimizing human bias and uncertainty. The workflow has been implemented in a Marcellus Shale reservoir with over 1500 wells to determine the optimal well spacing and completion design parameters for both the field and each well. The results show that using a shorter stage length and a higher sand-to-water ratio is preferable for this field, as it can increase cumulative gas production by up to 8%. Additionally, it is observed that fifty-percentile cumulative gas predictions are in close agreement with actual field productions. Thematic collection: This article is part of the Digitally enabled geoscience workflows: unlocking the power of our data collection available at: https://www.lyellcollection.org/topic/collections/digitally-enabled-geoscience-workflows

Methods for controlling heavy metals in environmental soils based on artificial neural networks

The problem of heavy metal pollution in soil has become a global environmental problem, and it is very important to predict and manage the heavy metals in the environmental soil in a timely manner. The changes in heavy metal content in soil have characteristics such as nonlinearity and large delay, making it difficult to predict heavy metals in soil using traditional methods. Traditional prediction methods are complex and cumbersome, which can lead to longer treatment time and easy secondary pollution. This article analyzed the Back Propagation neural network (BPNN) in artificial neural networks (ANN) and applied it to the prediction of heavy metals in environmental soils. BPNN has good nonlinear function approximation ability, so it can be well applied to complex problems such as soil heavy metal prediction. The methods of treating soil heavy metals include physical repair method, chemical repair method, microbial repair method, plant repair method, plant microbial combined repair method and so on. The use of BPNN can predict heavy metals in environmental soils through adaptive dynamic learning. However, the training time of the BPNN is relatively long and the convergence speed is relatively slow. Therefore, additional momentum terms were added to adjust the weights and thresholds of the network to improve the BPNN. In the experiment, the prediction performance of the improved BPNN was compared before and after the improvement. This article took 50 monitoring data of heavy metals in the same soil in a certain region in 2021 as sample data and predicted the content of heavy metals in the soil using improved and improved BPNN. Due to time constraints, this article only conducted experimental analysis on heavy metals such as lead and cadmium. In the first experiment, when the soil sample data was 50, the prediction accuracy of the BPNN for cadmium before and after improvement was 75.95% and 89.56%, respectively. In the second experiment, when the soil sample data was 50, the prediction accuracy of the BPNN for cadmium before and after improvement was 77.99% and 89.85%, respectively. The improved BPNN has good prediction accuracy and can effectively predict the status of heavy metals in soil. The analysis in this article can provide scientific basis for the comprehensive prevention and control of heavy metals in regional soil, and also provide reference for the development of pollution-free agriculture and ensuring food safety.

Neglected Cervical Facet Dislocation without Neurological Deficit: A Simple Solution to a Complex Problem—A Case Series and Review

AbstractCervical facet dislocation is a disastrous injury. Delayed presentation is not uncommon owing to various reasons in developing countries. Autofusion and a halt in progression are seen in spondylolysis and spondylolisthesis. Conservative treatment with close observation can be offered to the carefully selected patients with neglected cervical facet dislocations with intact neurology, as the injury is stable after autofusion. We registered eight male patients with cervical facet dislocation with delayed presentation for various reasons. Dynamic radiographs, computed tomography (CT) scan, and magnetic resonance imaging (MRI) were done to rule out any instability, cord compression, or bony fusion. Patients were followed up for 6 months clinically and radiologically. A total of eight patients presented with cervical facet dislocation, with a mean delay in presentation of 33 weeks (range: 14–54 weeks). Postinjury, all the patients were neurologically intact, with the only complaint of cervical pain at presentation. Dynamic radiographs did not show any instability. No cord compression or changes were noted on the MRI. Conservative management was advised for all eight patients. A subsequent follow-up CT scan demonstrated bony fusion at the dislocated facets in three of these patients. All the patients remained neurologically intact, with no further progression of the deformity. Surgical management is the mainstay of treatment for neglected facet dislocations. A combined approach has been advocated in most of the studies for proper reduction and alignment in these cases, which is quite extensive and has its own morbidity. A trial of conservative treatment can be offered to selected patients with intact neurology.

Network structure shapes the impact of diversity in collective learning

It is widely believed that diversity arising from different skills enhances the performance of teams, and in particular, their ability to learn and innovate. However, diversity has also been associated with negative effects on the communication and coordination within collectives. Yet, despite the importance of diversity as a concept, we still lack a mechanistic understanding of how its impact is shaped by the underlying social network. To fill this gap, we model skill diversity within a simple model of collective learning and show that its effect on collective performance differs depending on the complexity of the task and the network density. In particular, we find that diversity consistently impairs performance in simple tasks. In contrast, in complex tasks, link density modifies the effect of diversity: while homogeneous populations outperform diverse ones in sparse networks, the opposite is true in dense networks, where diversity boosts collective performance. Our findings also provide insight on how to forge teams in an increasingly interconnected world: the more we are connected, the more we can benefit from diversity to solve complex problems.

ELV Policy and Management System in Malaysia: Exploratory Study of Dynamics, Rationale and Opportunity through Digitalised Public Services

A sustainable end-of-life vehicle (ELV) management system requires a well-thought and well-designed process covering strategies, regulations, technologies, and future adaptive systems. Among others is the urgent need for feasible implementation of the integrated vehicle deregistration process, especially to kickstart the process for registered abandoned vehicles (approximately eight million units nationwide in 2018). In line with this aspiration, this study will explore the dynamics, rationale, and opportunity for public services to have a unified digitisation platform. This initiative will integrate networks of public (e.g., Ministry of Transport, Ministry of Housing and Local Government, Ministry of Environment and Water, Local councils, Road and Transportation Department, courts, Royal Malaysian Police), industry (e.g., authorised automotive treatment facilities (AATF), scrapyards) and other related parties. Those entities are critical to developing local ELV management system capacity to address current complex problems and achieve collective goals for the future automotive industry in Malaysia. Based on the Focus Group Discussion (FGD) outputs, literature survey, and relevant proposed frameworks, this study aims to enlighten policymakers on the necessity of having an integrated system in supporting various business, legal and administrative needs of the sustainable ELV Policy in Malaysia.

Determination of application boundaries to Giulotto method to during measuring of longitudinal relaxation time in nuclear magnetic flowmeters-relaxometers

The method of nuclear magnetic resonance, which is currently the most common in research and control of parameters of condensed matter, and the nuclear magnetic flowmeters-relaxometers that implement this method are described. The features of determining the times of longitudinal and transverse relaxation in nuclear magnetic flowmeters-relaxometers under different flow regimes of the medium in the process of monitoring its parameters have been established. The advantages of using a modulation technique for recording nuclear magnetic signals in flowmeters-relaxometers in comparison with other registration methods are noted. Using various approximations by the Giulotto method from the Bloch equations, a relation was obtained to determine the longitudinal relaxation time from the results of two measurements of the amplitudes of the nuclear magnetic resonance signal or resonant frequencies at different modulation frequencies. It has been experimentally proven that this relationship has a number of restrictions on its application for flowing liquid. These limitations are associated both with the technique of recording nuclear magnetic resonance signals and with the ability to generate such signals at different modulation frequencies of a constant magnetic field, the amplitudes of which differ from each other beyond the measurement error. The reasons that led to this discrepancy in the ratio for determining the time of longitudinal relaxation have been established. The limits of applicability of the obtained relation are found and it is experimentally proven that within these limits this relation can be used for reliable measurements of relaxation constants. Using experimental data, the relationship for determining the longitudinal relaxation time was studied. It has been proven that in a number of cases it is impossible to determine the desired value using the indicated relationship, although nuclear magnetic resonance signals of the current medium are recorded, and the medium has relaxation times. The results obtained make it possible to eliminate errors when using the nuclear magnetic resonance method to study flowing media and solve a number of complex problems in the energy, oil, chemical and pharmaceutical industries.

Stakeholder perceptions of gaps and solutions in addressing child trafficking

IntroductionGovernments, researchers, and community-based organizations have mobilized to develop interventions and programs addressing child trafficking, resulting in large amounts of financial resources allocated to awareness-raising campaigns, crisis services for survivors, and increasing judicial responses to trafficking. Low-resource and low-income countries face the dual challenge of addressing child trafficking within an environment where extreme poverty is pervasive and customary norms might conflict with national anti-trafficking policies. The purpose of this study is to understand challenges faced in addressing child trafficking and identify interventions that are community-driven and contextually relevant. MethodologyThe study used a qualitative community-based participatory research approach by collaborating with local researchers, government, and NGOs to examine how key stakeholders perceive existing gaps in policies and programs aimed at countering child trafficking in Sierra Leone. ResultsSix gaps in policies and services were identified by stakeholders: access to and availability of local programs and services for families; access to and affordability of schooling and vocational training in local communities; lack of policy implementation and enforcement of existing laws, especially at the local level; inadequate funding for anti-trafficking service providers; lack of community support for survivors of trafficking, and a lack of coordinated and consistent community response. In a demonstration of the trauma-informed and survivor-centered study design, participants also suggested possible ways to bridge these gaps. Discussion and conclusionAn ecological approach is critical in designing policies and programs responsive to the complex problem of child trafficking within an environment of low resources and potential conflicts between cultural norms and national policies. Practitioners in such contexts must be culturally-competent and prepared to seamlessly work across the micro-mezzo-macro continuum to enhance the best interest of the child.

Improved cooperative Ant Colony Optimization for the solution of binary combinatorial optimization applications

AbstractBinary combinatorial optimization plays a crucial role in various scientific and engineering fields. While deterministic approaches have traditionally been used to solve these problems, stochastic methods, particularly metaheuristics, have gained popularity in recent years for efficiently handling large problem instances. Ant Colony Optimization (ACO) is among the most successful metaheuristics and is frequently employed in non‐binary combinatorial problems due to its adaptability. Although for binary combinatorial problems ACO can suffer from issues such as rapid convergence to local minima, its eminently parallel structure means that it can be exploited to solve large and complex problems also in this field. In order to provide a versatile ACO implementation that achieves competitive results across a wide range of binary combinatorial optimization problems, we introduce a parallel multicolony strategy with an improved cooperation scheme to maintain search diversity. We evaluate our proposal (Binary Parallel Cooperative ACO, BiPCACO) using a comprehensive benchmark framework, showcasing its performance and, most importantly, its flexibility as a successful all‐purpose solver for binary combinatorial problems.

Information technology of synthesis of liquid dosing process control

Problem. Currently, different solutions are available on the global market for liquid dosing. There is no universal technology, as the same substances can be dosed using different types of equipment. Therefore, manufacturers face the complex task of determining the most optimal kind of equipment to achieve their goals. These issues are a complex multi-criteria scientific problem, and to effectively address them, it is necessary to develop a technology for synthesizing a liquid dosing control system using multi-criteria optimization methods in different conditions of information determinacy. Goal. The paper aims to increase the efficiency of liquid dosing control by developing information technology for the control system synthesizing. The object of the paper is the process of information technology development for synthesizing the liquid dosing control system. The subject of the paper is the models for the selection of hardware and software for the liquid dosing control system. Methodology. The study analyses and provides a classification of liquid dosing technology and analyses existing liquid dosing control systems. A structural model of information technology for synthesising a liquid dosing control system is developed, and criteria for selecting the elements of the liquid dosing control system are justified. A model for selecting PLC and SCADA systems for synthesising the liquid dosing control system is also developed using hierarchical analysis methods and multi-criteria evaluation and optimisation. Results. The paper resulted in the following: analysis and classification of liquid dosing technology; analysis of existing liquid dosing control systems; development of a structural model of the information technology of synthesising a liquid dosing control system; justification of criteria for selecting the elements of the liquid dosing control system, and the development of a model for selecting PLC and SCADA systems for the synthesis of the liquid dosing control system using hierarchical analysis methods and multi-criteria evaluation and optimisation. Originality. Further development of the methods of hierarchy analysis and multi-criteria assessment and optimization has been achieved by applying them to a specific field – the development of models for selecting hardware and software for the liquid dosing process control system. Practical value. The proposed information technology of synthesising a liquid dosing control system, as well as the mathematical models for selecting hardware and software for the liquid dosing process control system, will reduce the time and cost required to develop such a system.

Author identification of literary works based on text analysis and deep learning

With the development of science, speech, picture, and other analysis, problems have been gradually better solved, but the study of Chinese text has been a complex problem to overcome. Chinese text analysis requires not only statistics but also semantic comprehension analysis. Different text types need other language style feature modeling to obtain good recognition results. In this study, we use the deep learning method to construct an automatic text feature extraction model and classify it with the author as a classification label. This study presents a literature author recognition model based on deep learning, which is mainly divided into three phases: text preprocessing, feature extraction, and classification. Each part consists of several small modules or steps. First, we input the corpus to Word2Vec to generate the new word vector. Then, the improved text feature extractor based on CNN and Attention extracts the text features and uses them as the input of the CNN convolution layer. After convolution, the text is combined with bits to get Window Feature Sequence. It is the text feature vector. Next, based on LSTM and Softmax classification output, Window Feature Sequence is used as the input of LSTM to obtain two one-dimensional vectors spliced by concatenate layer. Finally, the result is classified through the fully connected layer, Batch Normalization layer, and Softmax. The performance of the proposed model in recognizing authors of Chinese literature was evaluated using two datasets. In the research process, the data we collected included works of different forms, such as prose and fiction. The research results show that the proposed model can effectively identify author identity. The classification accuracy of our proposed algorithm is significantly better than that of the benchmark model.

Helping students classify and frame capstone geotechnical design courses

Students often express an anxiety about how knowledge is classified (i.e., differentiated) and framed (i.e., prioritized, and sequenced) in capstone design problems. This anxiety is by design as capstone design courses are meant to test students’ ability to solve complex problems that are weakly classified and framed. Nevertheless, educators can play a role in scaffolding student progress, so students advance past a conceptual understanding of problems to applying technical acumen learnt in prior years. This paper presents three geotechnical design projects set by the author, along with three interventions used to scaffold student progress. Projects included the design of an industrial waste facility for dry filtered residue, design of remedial works for a clay river embankment subject to undercutting, and design of a remining method for mine slimes contained behind a sand embankment. Interventions included requiring students to prepare, present and critique presentations based on weekly stage gates, collaboratively brainstorming, and ranking high level implications of a design, and collaboratively brainstorming specific implications of a design. When implementing such interventions care must be taken to ensure they remain student driven, or the learning benefits of a capstone design course may be lost.

Antioxidant Activity and Bioactive Compounds of Fruits Available in Bangladesh

Background: The complex problems of different chronic degenerative diseases are spreading in human health across the glove. To combat this problem, fruits are the effective gear proved by many epidemiological studies. The aim of the study was to determine the antioxidant activity and three bioactive compounds like total flavonoids, total polyphenols and total anthocyanins in 23 species of fruits available in Bangladeshi market. Methods: Antioxidant activity (AOA) and bioactive compounds of fruits were determined by UV-visible spectroscopy. Antioxidant activity was undertaken by DPPH scavenging assay, total polyphenol content (TPC) was measured by folin-ciocalteu method and total flavonoid content (TFC) content was measured by aluminium chloride colorimetric method. Results: Among 23 species of fruits, AOA was higher in Katajamin (303 mg/100 g) followed by Dragon Fruits (171 mg/100 g), Guava (131 mg/100 g), Bael (130 mg/100 g) and average value was 125 mg/100 g of extract. TPC was varied from 10 to 305 mg GAE/100 g and average value was 87 mg GAE/100 g. Mango (305 mg GAE/100 g) showed higher TPC, followed by Pome Granate (222 mg GAE/100 g), Apple (207 mg GAE/100 g), Katajamin (141 mg GAE/100 g) and Bael (130 mg GAE/100 g). Again, TFC was ranged from Ashari (06±3.94 mg QE/g) to Tomato (84±13.69 mg QE/g) and the average TFC was 37 mg QE/g. In addition, total anthocyanin content (TAC) was highest in Mango (142±3.46 mg TA/100 g) and lower in Ashari (02±0.74 mg TA/100 g). There was a significant correlation (p≤0.05) between antioxidant activity and bioactive compounds and among the bioactive compounds. Conclusion: Important messages from this study are, majority of the rare fruits are the important sources of antioxidant and bioactive compounds which helps to reduce the risk of chronic degenerative diseases and can be used in food processing industry for nutraceutical product development.

A MODERN VIEW OF THE PROBLEM OF STRESS-INDUCED FERTILITY

Prolonged infertility can cause psychological stress in spouses, which requires the intervention of psychologists, psychotherapists, and sometimes psychiatrists. Purpose: to find out the role of stress in the initiation of infertility in a married couple. Materials and method: to conduct an analysis of modern medical literature on the psychogenic form of infertility. Research results and their discussion
 There are data that thanks to the attention and sensitive attitude to the mental state of patients, pregnancy occurred in 60% of cases. Can this be called a placebo reaction?
 Such a reaction can occur in the initial period of examination of women with probable anovulatory infertility. Psychological examination of a married couple with infertility should not be considered as an examination option, which is resorted to only after the exclusion of organic damage. Psychological and social factors can play a significant role when a woman has pathology of the pelvic organs. To assess the prospects of infertility treatment, the doctor must determine which of the factors - somatic, psychological or social - plays a leading role.
 Only then can he take the right position in relation to patients and assess the degree of his intervention in complex human problems that may arise in the process of diagnosis and treatment, or when choosing an alternative to parenthood.

Improving neural machine translation for low resource languages through non-parallel corpora: a case study of Egyptian dialect to modern standard Arabic translation

Machine translation for low-resource languages poses significant challenges, primarily due to the limited availability of data. In recent years, unsupervised learning has emerged as a promising approach to overcome this issue by aiming to learn translations between languages without depending on parallel data. A wide range of methods have been proposed in the literature to address this complex problem. This paper presents an in-depth investigation of semi-supervised neural machine translation specifically focusing on translating Arabic dialects, particularly Egyptian, to Modern Standard Arabic. The study employs two distinct datasets: one parallel dataset containing aligned sentences in both dialects, and a monolingual dataset where the source dialect is not directly connected to the target language in the training data. Three different translation systems are explored in this study. The first is an attention-based sequence-to-sequence model that benefits from the shared vocabulary between the Egyptian dialect and Modern Arabic to learn word embeddings. The second is an unsupervised transformer model that depends solely on monolingual data, without any parallel data. The third system starts with the parallel dataset for an initial supervised learning phase and then incorporates the monolingual data during the training process.

Where is Homelessness? When is Homelessness? Chronotopic Analysis of OECD Narratives of the Homelessness through Space, Time, and Body

Defining homelessness clearly without reducing the problem's complexity helps governments frame effective and conscious policies. There is a growing need for a theoretical framework that explores the common ground and generative structure among broad narratives about homelessness. In this article, we propose that Bakhtin’s (1981) concept of chronotope has excellent potential to achieve this goal. Chronotopes help us understand how time, space, and body configurations are represented in language and discourse for recognizing various situations and personas. Chronotope also enables us to reveal the assumptions and perspectives behind the narrative. Using a chronotopic lens, we analyzed narrations of homelessness from national governments and international organizations of 38 Organization for Economic Co-operation and Development (OECD) member countries. The findings reveal the dynamism, diversity, and assortment of dimensions related to time, space, and body in the analyzed narratives and highlight five main perspectives viewing homelessness as an issue of security, vulnerability, quantification, inclusivity, and human rights. The results of this analysis will guide upcoming research in two primary areas: investigating varied measurement methodologies for assessing different aspects of time, space, and body and scrutinizing how distinct attitudes towards homelessness impact policymaking and development processes.

Multi-beam line observation optimization strategy based on random variable unitary iterative model

Multibeam system has many advantages such as large measurement range, fast speed, high accuracy, etc., which can be applied in many fields such as marine topographic survey. In this paper, the relationship between coverage width and other variables is obtained by analyzing the geometric relationship between each parameter, and a multibeam sounding model based on geometric relationship is established. And based on the intelligent optimization algorithm, an iterative optimization model with random variable singularization is designed independently. And the actual terrain data were visualized and simplified into various parameters in line with the model. With the help of the definition of the coverage rate, the relationship between the coverage width and other variables, and the one-to-one correspondence between the translation distance of the slope and the coverage rate, the complex measurement problem can be transformed into a simple linear programming problem, and the relationship model between the coverage rate and other variables is established. In the end, the model can be used to derive the optimal wiring scheme to realize high-precision and full-coverage sweeping of the seabed terrain in the given sea area without omission, provided that the ship speed, beam opening angle and other factors are reasonable.

The Impact of the Employee’s Personal Characteristics on the Abuse of Sickness Absence: Empirical Evidence From Poland

Abstract Objective The tendency to abuse sickness absence is a complex problem conditioned by many factors, that can be divided, depending on the level of impact, into three groups: micro (related to personal characteristics), meso (related to the work environment), and macro (related to the broader environment). The subject of interest in the article is the first level (micro). The aim is to evaluate the impact of various personal factors on the declared abuse of sickness leave in Poland. Research Design & Methods The data was gathered using the CAWI survey. The research sample consisted of 1067 respondents (full-time employees). Statistical analysis incorporated multivariable linear regression and structural equation modelling. Findings Some personal characteristics have a significant impact on the declared abuse of sick leave. These factors are: gender, age, number of children, place of living, marital status, and subjective health. We found no impact of education and financial status. Implications & Recommendations Although abusing sick leave is a serious economic and organisational problem, its causes still remain relatively poorly recognised. Personal characteristics determine the propensity to abuse sick leave, and they need to be taken into account under the process of managing absence behaviours. Contribution & Value Added The added value of the study is drawing attention to the new personal factors which were not previously identified as a determinant of abuse of sick leave. These are: the number of children, place of living, and marital status.

THE BISECTOR TRIANGLE WITH A 90° ANGLE

For the examination of the triangle discussed in the article, profound knowledge of plane geometry is required, and grasping the specific approach to solving it can be quite challenging without prior knowledge of what to rely on. The focus is on a triangle with a 120° angle, commonly known as the Shebarshin problem. The article presents both computational and purely geometric methods for solving a problem of elevated complexity. This particular problem appears in entrance exams for universities and in mathematical Olympiads at the highest levels. The theme of singular triangles remains constantly relevant in geometry. For instance, many properties and theorems are known about an orthocentric triangle, which have been discussed in detail and already proven. I propose to consider in detail the properties and theorems of an equally interesting triangle, the sides of which connect the bases of the bisectors of the interior angles of the triangle. The goal of my article is to alleviate the drawback of complexity while retaining accessibility and originality in the geometric problem involving a triangle with a 120° angle and an inscribed bisectral triangle with a corresponding 90° angle.

Explainable Artificial Intelligence for Deep-Learning Based Classification of Cystic Fibrosis Lung Changes in MRI.

Algorithms increasing the transparence and explain ability of neural networks are gaining more popularity. Applying them to custom neural network architectures and complex medical problems remains challenging. In this work, several algorithms such as integrated gradients and grad came were used to generate additional explainable outputs for the classification of lung perfusion changes and mucus plugging in cystic fibrosis patients on MRI. The algorithms are applied on top of an already existing deep learning-based classification pipeline. From six explain ability algorithms, four were implemented successfully and one yielded satisfactory results which might provide support to the radiologist. It was evident, that the areas relevant for the classification were highlighted, thus emphasizing the applicability of deep learning for classification of lung changes in CF patients. Using explainable concepts with deep learning could improve confidence of clinicians towards deep learning and introduction of more diagnostic decision support systems.