CHRONIC KIDNEY DISEASE PREDICTION MODEL USING BAYESIAN OPTIMIZATION AND XGBOOST MACHINE LEARNING ALGORITHM

Unleashing the Power of Renal Nutrition in Value-Based Models of Kidney Care Choices: Leveraging Dietitians’ Expertise and Medical Nutrition Therapy to Delay Dialysis Initiation

The Third World Kidney Day: Looking Back and Thinking Forward

Background: Chronic kidney disease (CKD) is defined as abnormalities of kidney structure and/or function persisting for a minimum of three months.Methods: An online cross-sectional study was conducted in the Kingdom of Saudi Arabia (KSA) between June and August 2022 to assess the knowledge, attitudes, practices, and barriers among family, internal medicine, and general physicians related to CKD screening, diagnosis, and management.Results: A total of 427 physicians were included in the study. The majority exhibited a strong grasp of the accurate definition of CKD (83%) and recognized common risk factors, such as diabetes (99%), drugs (95%), and hypertension (98%). Two-thirds of physicians were aware of the five stages of CKD and identified estimated glomerular filtration rate (eGFR) and creatinine clearance as the most suitable markers for kidney function. Physicians also displayed knowledge of CKD-related complications. However, a noticeable gap between knowledge and practice was evident. Only one-third of participants reported screening their patients every year, primarily using serum creatinine (92.5%) and eGFR (97%) for diagnosis, while only 16% recognized that stage 4 CKD is the appropriate time to refer patients to nephrologists. In terms of barriers, the majority of physicians reported encountering low barriers to CKD management, but general practitioners working in primary healthcare centers experienced significantly higher levels of barriers.Conclusion: Most participants in our study possess a good level of knowledge and positive attitudes towards CKD diagnosis and management. Nevertheless, a discrepancy between knowledge and practical application, particularly in terms of over-screening and early referral, highlights the need for educational efforts to improve physician practice in KSA. These findings underscore the importance of addressing this gap to promote effective CKD management.

IntroductionThe burden of chronic kidney disease (CKD) is rising globally including in Singapore. Primary care is the first point of contact for most patients with early stages of CKD. However, several barriers to optimal CKD management exist. Knowing healthcare professionals’ (HCPs) perspectives is important to understand how best to strengthen CKD services in the primary care setting. Integrating a theory-based framework, we explored HCPs’ perspectives on the facilitators of and barriers to CKD management in primary care clinics in Singapore.MethodsIn-depth interviews were conducted on a purposive sample of 20 HCPs including 13 physicians, 2 nurses and 1 pharmacist from three public primary care polyclinics, and 4 nephrologists from one referral hospital. Interviews were audio recorded, transcribed verbatim and thematically analyzed underpinned by the Theoretical Domains Framework (TDF) version 2.ResultsNumerous facilitators of and barriers to CKD management identified. HCPs perceived insufficient attention is given to CKD in primary care and highlighted several barriers including knowledge and practice gaps, ineffective CKD diagnosis disclosure, limitations in decision-making for nephrology referrals, consultation time, suboptimal care coordination, and lack of CKD awareness and self-management skills among patients. Nevertheless, intensive CKD training of primary care physicians, structured CKD-care pathways, multidisciplinary team-based care, and prioritizing nephrology referrals with risk-based assessment were key facilitators. Participants underscored the importance of improving awareness and self-management skills among patients. Primary care providers expressed willingness to manage early-stage CKD as a collaborative care model with nephrologists. Our findings provide valuable insights to design targeted interventions to enhance CKD management in primary care in Singapore that may be relevant to other countries.ConclusionsThe are several roadblocks to improving CKD management in primary care settings warranting urgent attention. Foremost, CKD deserves greater priority from HCPs and health planners. Multipronged approaches should urgently address gaps in care coordination, patient-physician communication, and knowledge. Strategies could focus on intensive CKD-oriented training for primary care physicians and building novel team-based care models integrating structured CKD management, risk-based nephrology referrals coupled with education and motivational counseling for patients. Such concerted efforts are likely to improve outcomes of patients with CKD and reduce the ESKD burden.

Diabetes mellitus (DM) is a primary cause of chronic kidney disease (CKD), a significant and growing global health issue. Primary healthcare physicians (PHPs) are crucial in detecting, managing, and preventing CKD, but many lack the necessary knowledge and skills to effectively diagnose and manage the disease. This study assessed the knowledge and competence of PHPs in managing diabetes-related CKD. This cross-sectional study was conducted among all Primary Healthcare Physicians (PHPs) working at Primary Healthcare Center (PHC) in King Abdulaziz Medical City of the National Guard, Jeddah, Saudi Arabia. An online questionnaire was sent to all PHPs and data were collected from January 2024 to February 2024. Data analysis performed utilizing RStudio (version 4.3.1). Categorical variables were expressed as frequencies and percentages, whereas mean and standard deviations were computed for continuous variables. Kruskal-Wallis rank sum test was used to determine differences in confidence scores across various demographic and occupational characteristics. A multivariable linear regression analysis was performed to identify factors related to confidence. Of 141 PHPs, 122 filled online questionnaire yielding a response rate of 86.5%. responded. The majority (90.2%) were 18-45 years of age, 68.0% were qualified as family physicians, 38.0% belonged to the specialized poly clinic, and 54.9% had less than 5 years of experience. PHPs showed a varied levels of confidence and knowledge. While 76.2% were confident about the stages of kidney disease and 65.6% acknowledged the importance of the urine albumin-creatinine ratio (uACR) test, 58.1% were uncertain of interpreting uACR results and CKD diagnostic criteria. More than 60.0% were uncertain of the treatment steps postdiagnosis and prediction of CKD prognosis. Age, qualification, specialization in family medicine, and clinic affiliation were significantly related to confidence levels. The study underscores notable gaps in PHPs' knowledge and confidence concerning CKD screening, diagnosis, and management. Continuous education and targeted interventions are essential for the improvement of PHPs' competence and patient outcomes in the management of CKD.

Protein metabolism is disturbed by chronic kidney disease (CKD), a progressive illness that raises the risk of cardiovascular disease and causes metabolic problems such as inflammation and protein-energy waste (PEW). Maintaining nutritional health, lowering uremia, and slowing the course of CKD all depend on proper dietary protein control. The impact of dietary changes, such as low-protein diets (LPD) and plant-based protein sources, and protein supplements on the management of CKD are examined in this study. Research indicates that although LPD aids in the management of metabolic issues, it necessitates close observation to avoid muscular atrophy and malnutrition. While consuming too much animal protein, especially red meat, can hasten the evolution of CKD, plant-based proteins assist the kidneys by lowering acid load, phosphorus bioavailability, and uremic toxin generation. Additionally, although more research is required, LPD may improve renal protection when used with renin-angiotensin-aldosterone system (RAAS) inhibitors. Specialised nutritional techniques are needed for the management of paediatric CKD in order to balance growth and avoid malnutrition. The study emphasises the value of tailored dietary therapies and the need for more research to improve long-term dietary plans and protein consumption guidelines for individuals with chronic kidney disease.

Chronic kidney disease (CKD) patients have a shorter life expectancy than healthy individuals, from the premature development of cardiovascular disease due to traditional and non-traditional risk factors. It is therefore pertinent that management of CKD be aimed at modifying these cardiovascular disease risk factors. Exercise therapy, in particular aerobic exercise, in the end stage kidney disease population has been shown to decrease cardiovascular disease risk factors and subsequent cardiovascular disease. However, randomized exercise training trials are scarce in the stage 3-4 CKD cohort. High intensity interval training (HIIT) is a promising exercise training approach that has been shown to elicit greater improvements in cardiovascular disease risk factors than moderate intensity continuous training (MICT) in other chronic disease populations. However, HIIT has not yet been investigated in patients with CKD. This thesis comprises two sections, each containing a randomised control trial (RCT) investigating the effects of exercise training in CKD patients. Section 1 contains data from the Landmark III (LM3) trial, which investigated the effects of a nurse-led, multi-disciplinary lifestyle intervention (LI) in 187 CKD patients. Testing of both LI and standard care (control) groups occurred every 6 months for 3 years. The exercise training in the LI consisted of a combination of individualized aerobic and resistance training and encouraged patients to participate in higher intensity exercise if possible. The main study in Section 1 assesses the change in various outcome measures after 3 years of LI (Chapter 6). Section 1 also includes baseline cross-sectional analyses (Chapter 3) and 12 month sub-studies (Chapters 4 and 5). Section 2 contains the second trial that investigated the feasibility and efficacy of a 12 week HIIT program in CKD patients. Recruitment for the HIIT study was through participants who had completed the LM3 study. Chapter 1 provides an introduction to the studies, including the aims of the thesis. Topics reviewed include; pathophysiology of CKD, exercise capacity, autonomic dysfunction, haemoglobin, muscle atrophy, oxidative stress, inflammation, blood lipids and HIIT. Section 1 starts with the detailed methods of the LM3 study (Chapter 2). Further methods applicable to specific studies are reported in the relevant chapter. Chapter 3 investigates the determinants of reduced muscle strength at baseline through a cross-sectional analysis of the LM3 study. This study identified the association of oxidative stress with low muscle strength and reduced lean mass. The relationship was independent of known influences on muscle mass and strength. The feasibility and determinants of the LI group participating in higher intensity exercise training in the first 12 months of the LM3 study was investigated in Chapter 4. The findings identified that higher intensity exercise training was indeed feasible in the CKD cohort, with the number of participants reporting higher intensity exercise increasing from 20% to 43%. Reported higher intensity exercise during the LI was also associated with the greatest exercise capacity at 12 months. Furthermore, the analysis identified haemoglobin to be the greatest predictor of individuals reporting higher intensity exercise. Chapter 5 investigates the agreement between creatinine and cystatin-C estimated glomerular filtration rate (eGFR) measurements in patients completing 12 months of the LM3 study. This chapter explores whether discrepancies in the CKD Epidemiology Collaboration (CKD-EPI)cystatin-C, CKD-EPIcreatinine and modification of diet in renal disease (MDRD) eGFR equations occur after an exercise intervention. It was identified that cystatin-C based eGFR measurements were considerably lower than creatinine based eGFR measurements. Despite this discrepancy at baseline, the agreement between the different GFR estimates did not change after exercise training. Chapter 6 contains the main study from LM3, which investigates the changes in cardiorespiratory fitness, exercise capacity, functional capacity, blood biochemistry, anthropometry measures and other cardiovascular disease risk factors after 3 years of LI compared to the control group. The primary finding from this study was the ability of a long-term LI to significantly increase physical activity levels, exercise capacity and autonomic function and decrease arterial stiffness in patients with CKD. Section 2 includes a systematic review and meta-analysis of HIIT in patients with cardiometabolic diseases (Chapter 7). This review investigated the efficacy and safety of HIIT compared to MICT in individuals with a population sample of chronic disease, where poor lifestyle was considered a main contributor to the disease. The meta-analysis was performed using oxygen uptake (VO2peak or VO2maximum) as the primary outcome measure, however the influence of HIIT on all cardiovascular disease risk factors was explored. It was identified that HIIT provokes nearly twice the increase in VO2peak compared to MICT. Chapter 8 in Section 2 investigates the feasibility of HIIT and the efficacy of HIIT compared to MICT in improving cardiovascular disease risk factors. It was identified that HIIT was a feasible option for patients with CKD. However, the improvements in cardiorespiratory fitness and exercise capacity were slightly greater in the MICT than the HIIT group, although the changes were not statistically significant. Participants from the HIIT group also reported higher levels of enjoyment from the intervention, as measured by the Physical Activity Enjoyment Scale. Chapter 9 provides a summary of the outcomes from all studies. This chapter discusses how this thesis has extensively contributed to the current knowledge in exercise training in CKD patients. Based on the findings, it is recommended that a combination of aerobic and resistance training, with some higher intensity exercise if possible, be prescribed to patients with CKD to elicit the greatest health outcomes. Important factors to consider with exercise training and testing of CKD patients, such as the effects of certain medications and hydration status on physiological responses, are also explored. Further, this chapter suggests future directions for research which will extend the findings from this thesis.

Background and Aims Chronic Kidney Disease (CKD) represents a significant public health challenge. The primary care physician holds a crucial position in facilitating its early detection, serving as a linchpin in the comprehensive management of patients often with multiple comorbidities. The objective of our study was to investigate CKD management in primary care practitioners in Tunisia and their relationship with nephrologists. Method We conducted a cross-sectional study involving family physicians and general practitioners. This involved a self-administered questionnaire using Google Forms, aiming to investigate CKD management in the context of primary care in Tunisia, along with an examination of collaboration with nephrologists. Results Our study included 43 practitioners with a gender ratio of M/F = 0.3. The average age of participants was 30.7 years [25-53 years]. Of the 43 participants, 95% practiced in urban areas, and 74% were in training. They worked in university hospitals (53%), primary care centers (21%), regional hospitals (19%), and private practices (7%). When asked about their feelings during the management of a patient with CKD, the main response was anxiety (65%) followed by indifference (18%), confidence (12%), and fear (5%). According to practitioners, the uniqueness of CKD management was due to the polypathological nature of patients (79%), the difficulty in managing certain cases (74%), the difficulty in applying certain recommendations (60%), the lack of medications (2%), and the difficulty in accessing complementary tests in primary care (2%). The means of training and sources of information on CKD mentioned were specialized medical journals (49%), medical websites (40%), peer groups (4%), pharmaceutical industry representatives (2%), and no source (5%). The means of training included additional studies certificates (23%), workshops/plenary sessions (44%), specialized webinars (30%), intra-hospital medical days (46%), nephrology internships (28%), and no available means (16%). Regarding the relationship with the nephrologist in joint management, 42% had a corresponding nephrologist. Thirty percent declared no difficulty in referring patients to nephrology when necessary. In case of difficulty, it would be related to long delays in obtaining a consultation with the nephrologist (60%), the geographical distance of nephrology centers (30%), and sometimes patient refusal (23%). When discovering CKD, 60% immediately referred patients to the nephrologist. On a scale of 0 to 5, communication and coordination of care between nephrologists and primary care physicians were estimated at 0 by 12%, 1 by 16%, 2 by 30%, 3 by 30%, 4 by 12%. Regarding end stage CKD management, practitioners informed patients about kidney transplantation (9%), hemodialysis (53%), peritoneal dialysis (16%), and 55% stated they did not have to inform patients about these techniques. Sixty-three percent continued to follow patients after the initiation of renal replacement therapy. Conclusion Our work revealed that the predominant sentiment when dealing with CKD patients in primary care is anxiety. This emotion appears to stem from the intricate nature of CKD management, primarily associated with the polypathological profile of patients and challenges in implementing recommended guidelines, compounded by difficulties in accessing nephrologist consultations. The imperative challenge lies in enhancing information accessibility for practitioners to optimize CKD management within this medical context.

Background: Patients with advanced chronic kidney disease (CKD) experience many bothersome symptoms, affecting their quality of life (QoL). The majority of these patients do not have the opportunity to receive a kidney transplant and need other treatments for these burdensome symptoms to improve their QoL and prognosis. With growing attention to the significance of symptom management in advanced CKD on haemodialysis (HD), the evidence regarding symptoms is increasing. Methods: In this review, we briefly summarise the current evidence of non-pharmacologic interventions to improve symptoms and QoL in patients with advanced CKD, on dialysis. Hence, we focused on symptoms that are commonly experienced by patients, such as pain, fatigue, sleep disturbances, itching, nausea and vomiting, cognitive impairment and anxiety and depression. Results: We noted that research on improving symptom management in CKD is still very limited. In addition to the lack of clinical practice guidelines to address those common symptoms, the evidence to how to incorporate effective symptom intervention approaches into clinical care is also rare. Conclusion: Although improving mortality remains an important area in the kidney community, there is an urgent need to focus on improving symptommanagement to improve QoL in advanced CKD on HD.

INTRODUCTION: Chronic kidney disease (CKD) has been recognized as a leading public health problem worldwide. Early detection and management of the risk factors of CKD is helpful for its prevention and reduction of burden caused by it. This study aimed to find out the awareness regarding chronic kidney disease among adults in a hospital of Siddharthanagar Municipality. MATERIAL AND METHODS: Hospital based descriptive cross-sectional study was used to find out the awareness regarding chronic kidney disease among 165 adults attending Medical out patient department of Universal College of Medical Sciences, Bhairahawa, Rupandehi, Nepal from July 11 to July 25, 2021. Descriptive and inferential statistics were used to assess the awareness level and its association with different socio-demographic variables. RESULTS: The findings of the study showed that 57.6% of respondents had low level of awareness regarding chronic kidney disease. Respondents have high awareness on risk factors of chronic kidney disease whereas respondents have relatively low awareness on meaning and clinical features of chronic kidney disease. CONCLUSION: On the basis of findings, it is concluded that more than half of the respondents have low awareness level regarding chronic kidney disease so an effort should be made to improve the awareness regarding chronic kidney disease for early identification and management of chronic kidney disease.

PurposeChronic diseases are considered as one of the serious concerns and threats to public health across the globe. Diseases such as chronic diabetes mellitus (CDM), cardio vasculardisease (CVD) and chronic kidney disease (CKD) are major chronic diseases responsible for millions of death. Each of these diseases is considered as a risk factor for the other two diseases. Therefore, noteworthy attention is being paid to reduce the risk of these diseases. A gigantic amount of medical data is generated in digital form from smart healthcare appliances in the current era. Although numerous machine learning (ML) algorithms are proposed for the early prediction of chronic diseases, these algorithmic models are neither generalized nor adaptive when the model is imposed on new disease datasets. Hence, these algorithms have to process a huge amount of disease data iteratively until the model converges. This limitation may make it difficult for ML models to fit and produce imprecise results. A single algorithm may not yield accurate results. Nonetheless, an ensemble of classifiers built from multiple models, that works based on a voting principle has been successfully applied to solve many classification tasks. The purpose of this paper is to make early prediction of chronic diseases using hybrid generative regression based deep intelligence network (HGRDIN) model.Design/methodology/approachIn the proposed paper generative regression (GR) model is used in combination with deep neural network (DNN) for the early prediction of chronic disease. The GR model will obtain prior knowledge about the labelled data by analyzing the correlation between features and class labels. Hence, the weight assignment process of DNN is influenced by the relationship between attributes rather than random assignment. The knowledge obtained through these processes is passed as input to the DNN network for further prediction. Since the inference about the input data instances is drawn at the DNN through the GR model, the model is named as hybrid generative regression-based deep intelligence network (HGRDIN).FindingsThe credibility of the implemented approach is rigorously validated using various parameters such as accuracy, precision, recall, F score and area under the curve (AUC) score. During the training phase, the proposed algorithm is constantly regularized using the elastic net regularization technique and also hyper-tuned using the various parameters such as momentum and learning rate to minimize the misprediction rate. The experimental results illustrate that the proposed approach predicted the chronic disease with a minimal error by avoiding the possible overfitting and local minima problems. The result obtained with the proposed approach is also compared with the various traditional approaches.Research limitations/implicationsUsually, the diagnostic data are multi-dimension in nature where the performance of the ML algorithm will degrade due to the data overfitting, curse of dimensionality issues. The result obtained through the experiment has achieved an average accuracy of 95%. Hence, analysis can be made further to improve predictive accuracy by overcoming the curse of dimensionality issues.Practical implicationsThe proposed ML model can mimic the behavior of the doctor's brain. These algorithms have the capability to replace clinical tasks. The accurate result obtained through the innovative algorithms can free the physician from the mundane care and practices so that the physician can focus more on the complex issues.Social implicationsUtilizing the proposed predictive model at the decision-making level for the early prediction of the disease is considered as a promising change towards the healthcare sector. The global burden of chronic disease can be reduced at an exceptional level through these approaches.Originality/valueIn the proposed HGRDIN model, the concept of transfer learning approach is used where the knowledge acquired through the GR process is applied on DNN that identified the possible relationship between the dependent and independent feature variables by mapping the chronic data instances to its corresponding target class before it is being passed as input to the DNN network. Hence, the result of the experiments illustrated that the proposed approach obtained superior performance in terms of various validation parameters than the existing conventional techniques.

CKD encompasses a spectrum of pathophysiologic processes associated with abnormal kidney function and a progressive decline in the glomerular filtration rate (GFR). Elimination of Malas from the body is also an inductive of good health. There are totally three Malas explained by the Samhitas namely Purisha, Mutra & Sweda. In Chronic Kidney Disease (CKD) where there is a less formation of Mutra, the Karma of Mutra is removing Kleda (waste products) from the body. So, the Kleda which resides in the body causes Pratiloma gati of Vata leading to different variety of diseases which involves Dusti of Rakta. Therefore, use of Mutrala & Raktashodhaka Dravyas may be helpful in the subjects of CKD. There is no availability of direct description of CKD in Ayurvedic science, except Vrukka roga Adhikara of Bhaishajya Ratnavali. So, we studied the disease with Ayurvedic concepts on the basis of general signs and symptoms.
 Here 28 subjects diagnosed with Chronic Kidney Disease (CKD) fulfilling the inclusion criteria were selected incidentally for study. For each subject of CKD Amapachana and Koshtashodhana was done with Hareetakyadi churna, Anubhuta Kashaya and Kaishora Guggulu are administered as Shamanoushadhi. With this intervention, we are able to give mild to moderate improvement in subjective and objective parameters. During the study improvement of subjective parameters are well appreciated than the objective parameters. The objective of the study is to establish the combined effectiveness of Anubhuta Kashaya and Kaishora guggulu in the management of Chronic Kidney Disease.

BackgroundChronic kidney disease (CKD) is associated with various health complications, including sleep disorders. Understanding the relationship between CKD and sleep disturbances is essential for improving patient management and outcomes in this population.Patients and methodsThis study analyzed data from the National Health and Nutrition Examination Survey (NHANES) across two cycles (2015–2016 and 2017-2018), involving 4758 participants aged 20 years and older. Sleep disorders were evaluated via targeted questions from the Sleep Disorders Questionnaire. Chronic kidney disease (CKD) was classified by low estimated glomerular filtration rate (eGFR) and proteinuria. Logistic regression analyses examined the association between sleep disorders and CKD, adjusting for demographic and lifestyle confounders.ResultsAmong 4,758 participants, 863 (14%) were diagnosed with chronic kidney disease (CKD). Elevated C-reactive protein (CRP) levels and differential leukocyte counts were observed in the CKD cohort. CKD was associated with increased reports of snoring, sleep apnea, and sleep disturbances, along with higher fatigue levels. A significant positive association between CKD and sleep difficulties persisted in adjusted models. Infrequent snoring (1–2 times per week) demonstrated a negative association, whereas excessive daytime sleepiness (16–30 times/month) correlated positively with CKD. Significant associations between sleep disorders and CKD were observed in participants aged <60, with a BMI <28 kg/m², normal blood pressure, and CRP ≥1.8 mg/L. Sleep disorders were found to be correlated with obesity, hypertension, diabetes, CKD, and proteinuria. Notably, CKD patients with sleep difficulties had markedly elevated CRP levels compared to those without sleep issues, while other inflammatory markers were similarly elevated (P < 0.001).ConclusionPatients with chronic kidney disease experience higher rates of sleep issues, highlighting the importance of addressing these problems in CKD management.

Chronic Kidney Disease (CKD) is a significant global health challenge, with increasing prevalence and a high burden of associated morbidity and mortality. Early detection and management of CKD are essential in preventing its progression to end-stage renal disease, which often requires dialysis or transplantation. Primary care settings play a pivotal role in the early identification and management of CKD, given their accessibility and the opportunity for preventative care. This article aims to identify and review the key risk factors for CKD in primary care populations. A comprehensive literature review was conducted using data from peer-reviewed articles, clinical guidelines, and epidemiological studies, with a focus on demographic, medical, lifestyle, and genetic risk factors. The review highlights hypertension, diabetes, obesity, family history, and ethnicity as major contributors to CKD risk. Additionally, modifiable factors such as diet, smoking, and physical inactivity are emphasized. The findings underscore the importance of regular screening and early intervention in primary care to mitigate CKD progression. Primary care practitioners must prioritize CKD risk assessment, raise awareness about preventive strategies, and manage comorbidities effectively. Addressing these risk factors at the primary care level could significantly reduce the overall burden of CKD, improve patient outcomes, and alleviate healthcare costs. Public health strategies must focus on promoting early detection and equitable access to care for at-risk populations.

Accurate and reliable lithological mapping through satellite-borne remote sensing data and image classification approaches has a critical role since it can automatically and promptly identify lithological units over large areas. Most available Pixel-Object Based comparative classification studies have been applied to land use land cover (LULC) studies; however, this research aims to evaluate and compare the performance of these digital classification methods in the field of geological mapping in semi-arid areas, by integrating spectral bands and neo-bands, particularly the Minimum noise fraction (MNF) and the principal component analysis (PCA), of Sentinel-2A satellite imagery, to map the southern of Skhour Rehamna which is located at the western Moroccan Meseta. The analysis results from two different methods, namely, pixel-based image analysis (PBIA) with k-nearest neighbour (K-NN) and Random Forest (RF) machine learning algorithms (MLAs), and Geographic Object-Based Image Analysis (GEOBIA) were assessed and compared. PBIA method involved selection of training areas whether it was k-NN or RF MLAs, and produced lithological maps that exhibit “salt and pepper” effects as well as problems associated to delineating accurate lithological boundaries, while GEOBIA approach involved multi-resolution segmentation step where scale, shape and compactness parameters should be adjusted as accurate as possible, in order to segment the image into homogeneous and meaningful regions so that the resulted samples were classified using Standard Nearest Neighbour algorithm. Therefore, the resulting lithological maps were assessed by comparing both techniques using confusion matrix, overall accuracy (OA) and Kappa coefficient (K). The results show that the GEOBIA approach had higher overall agreement (83.46% OA and 0.76 K) than RF (81.92% OA and 0.72 K) and k-NN (80.79% OA and 0.70 K) PBIA approaches. Overall, the results clearly indicate the potential of GEOBIA technique for lithological mapping applications to produce more realistic maps.