1.Feature Selection for Hypertension Risk Prediction Using XGBoost on Single Nucleotide Polymorphism Data
Lailil MUFLIKHAH ; Tirana Noor FATYANOSA ; Nashi WIDODO ; Rizal Setya PERDANA ; Solimun ; Hana RATNAWATI
Healthcare Informatics Research 2025;31(1):16-22
Objectives:
Hypertension, commonly known as high blood pressure, is a prevalent and serious condition affecting a significant portion of the adult population globally. It is a chronic medical issue that, if left unaddressed, can lead to severe health complications, including kidney problems, heart disease, and stroke. This study aims to develop a feature selection model using the XGBoost algorithm to identify specific single nucleotide polymorphisms (SNPs) as biomarkers for detecting hypertension risk.
Methods:
We propose using the high dimensionality of genetic variations (i.e., SNPs) to build a classifier model for prediction. In this study, SNPs were used as markers for hypertension in patients. We utilized the OpenSNP dataset, which includes 19,697 SNPs from 2,052 samples. Extreme gradient boosting (XGBoost) is an ensemble machine learning method employed here for feature selection, which incrementally adjusts weights in a series of steps.
Results:
The experimental results identified 292 SNPs that exhibited high performance, with an F1-score of 98.55%, precision of 98.73%, recall of 98.38%, and overall accuracy of 98%. This study provides compelling evidence that the XGBoost feature selection method outperforms other representative feature selection methods, such as genetic algorithms, analysis of variance, chi-square, and principal component analysis, in predicting hypertension risk, demonstrating its effectiveness.
Conclusions
We developed a model for predicting hypertension using the SNPs dataset. The high dimensionality of SNP data was effectively managed to identify significant features as biomarkers using the XGBoost feature selection method. The results indicate high performance in predicting the risk of hypertension.
2.Pseudomonas aeruginosa in Chronic Suppurative Otitis Media
Artono ARTONO ; Nyilo PURNAMI ; Edi HANDOKO ; Agung Dwi Wahyu WIDODO ; Juniastuti JUNIASTUTI
Infection and Chemotherapy 2025;57(1):63-71
Background:
Chronic suppurative otitis media (CSOM) has caused many hearing disorder cases in developing countries.Inappropriate antibiotic use resulted in a shift of bacterial resistance. The biofilm-forming bacteria, like Pseudomonas aeruginosa, was a common germ detected in CSOM that contributed to a poor prognosis. This study aimed to investigate the bacterial pattern from samples taken from CSOM patients regarding its antibiotic susceptibility and the antibiofilm activity of acetic acid against P. aeruginosa.
Materials and Methods:
Sterile swabs of forty-five patients with CSOM were collected, followed by isolation of bacterial pathogens, identification, and evaluation of antibiotic sensitivity using modified Kirby Bauer disc diffusion protocol. In vitro testing was done by adding acetic acid to P. aeruginosa culture to gauge the minimum concentration of biofilm inhibition and eradication. They were conducted using the microtiter plate assay method and quantified with an ELISA reader. The data were analyzed statistically using One-Way ANOVA and Tukey Honestly Significant Difference post hoc test.
Results:
The samples obtained from 31 of 45 CSOM patients showed positive microbial growth; 26 (57.78%) had a monomicrobial pattern, and 5 (11.11%) had a polymicrobial pattern. The researcher ascertained that 24 isolates, representing 66.67%, were gram-negative bacteria, with P.-aeruginosa identified as the predominant species.P. aeruginosa isolates were sensitive to several antibiotics, including meropenem, amikacin, piperacillin-tazobactam, ceftazidime, and cefoperazone-sulbactam with a rate of 93.33%. The minimum concentration of acetic acid required to qualify as the minimum biofilm inhibitory concentration (MBIC) was determined to be 0.16%, yielding an inhibition rate of 26.79%. A concentration of 0.31% was identified as the minimum biofilm eradication concentration (MBEC), achieving an eradication rate of 77.27%.
Conclusion
P. aeruginosa, the most common bacteria found in CSOM samples, was sensitive to imipenem, amikacin, piperacillin-tazobactam, ceftazidime, and cefoperazone-sulbactam. Acetic acid suppresses P. aeruginosa bacterial biofilm formation at MBIC of 0.16% and MBEC of 0.31%.
3.Feature Selection for Hypertension Risk Prediction Using XGBoost on Single Nucleotide Polymorphism Data
Lailil MUFLIKHAH ; Tirana Noor FATYANOSA ; Nashi WIDODO ; Rizal Setya PERDANA ; Solimun ; Hana RATNAWATI
Healthcare Informatics Research 2025;31(1):16-22
Objectives:
Hypertension, commonly known as high blood pressure, is a prevalent and serious condition affecting a significant portion of the adult population globally. It is a chronic medical issue that, if left unaddressed, can lead to severe health complications, including kidney problems, heart disease, and stroke. This study aims to develop a feature selection model using the XGBoost algorithm to identify specific single nucleotide polymorphisms (SNPs) as biomarkers for detecting hypertension risk.
Methods:
We propose using the high dimensionality of genetic variations (i.e., SNPs) to build a classifier model for prediction. In this study, SNPs were used as markers for hypertension in patients. We utilized the OpenSNP dataset, which includes 19,697 SNPs from 2,052 samples. Extreme gradient boosting (XGBoost) is an ensemble machine learning method employed here for feature selection, which incrementally adjusts weights in a series of steps.
Results:
The experimental results identified 292 SNPs that exhibited high performance, with an F1-score of 98.55%, precision of 98.73%, recall of 98.38%, and overall accuracy of 98%. This study provides compelling evidence that the XGBoost feature selection method outperforms other representative feature selection methods, such as genetic algorithms, analysis of variance, chi-square, and principal component analysis, in predicting hypertension risk, demonstrating its effectiveness.
Conclusions
We developed a model for predicting hypertension using the SNPs dataset. The high dimensionality of SNP data was effectively managed to identify significant features as biomarkers using the XGBoost feature selection method. The results indicate high performance in predicting the risk of hypertension.
4.Pseudomonas aeruginosa in Chronic Suppurative Otitis Media
Artono ARTONO ; Nyilo PURNAMI ; Edi HANDOKO ; Agung Dwi Wahyu WIDODO ; Juniastuti JUNIASTUTI
Infection and Chemotherapy 2025;57(1):63-71
Background:
Chronic suppurative otitis media (CSOM) has caused many hearing disorder cases in developing countries.Inappropriate antibiotic use resulted in a shift of bacterial resistance. The biofilm-forming bacteria, like Pseudomonas aeruginosa, was a common germ detected in CSOM that contributed to a poor prognosis. This study aimed to investigate the bacterial pattern from samples taken from CSOM patients regarding its antibiotic susceptibility and the antibiofilm activity of acetic acid against P. aeruginosa.
Materials and Methods:
Sterile swabs of forty-five patients with CSOM were collected, followed by isolation of bacterial pathogens, identification, and evaluation of antibiotic sensitivity using modified Kirby Bauer disc diffusion protocol. In vitro testing was done by adding acetic acid to P. aeruginosa culture to gauge the minimum concentration of biofilm inhibition and eradication. They were conducted using the microtiter plate assay method and quantified with an ELISA reader. The data were analyzed statistically using One-Way ANOVA and Tukey Honestly Significant Difference post hoc test.
Results:
The samples obtained from 31 of 45 CSOM patients showed positive microbial growth; 26 (57.78%) had a monomicrobial pattern, and 5 (11.11%) had a polymicrobial pattern. The researcher ascertained that 24 isolates, representing 66.67%, were gram-negative bacteria, with P.-aeruginosa identified as the predominant species.P. aeruginosa isolates were sensitive to several antibiotics, including meropenem, amikacin, piperacillin-tazobactam, ceftazidime, and cefoperazone-sulbactam with a rate of 93.33%. The minimum concentration of acetic acid required to qualify as the minimum biofilm inhibitory concentration (MBIC) was determined to be 0.16%, yielding an inhibition rate of 26.79%. A concentration of 0.31% was identified as the minimum biofilm eradication concentration (MBEC), achieving an eradication rate of 77.27%.
Conclusion
P. aeruginosa, the most common bacteria found in CSOM samples, was sensitive to imipenem, amikacin, piperacillin-tazobactam, ceftazidime, and cefoperazone-sulbactam. Acetic acid suppresses P. aeruginosa bacterial biofilm formation at MBIC of 0.16% and MBEC of 0.31%.
5.Feature Selection for Hypertension Risk Prediction Using XGBoost on Single Nucleotide Polymorphism Data
Lailil MUFLIKHAH ; Tirana Noor FATYANOSA ; Nashi WIDODO ; Rizal Setya PERDANA ; Solimun ; Hana RATNAWATI
Healthcare Informatics Research 2025;31(1):16-22
Objectives:
Hypertension, commonly known as high blood pressure, is a prevalent and serious condition affecting a significant portion of the adult population globally. It is a chronic medical issue that, if left unaddressed, can lead to severe health complications, including kidney problems, heart disease, and stroke. This study aims to develop a feature selection model using the XGBoost algorithm to identify specific single nucleotide polymorphisms (SNPs) as biomarkers for detecting hypertension risk.
Methods:
We propose using the high dimensionality of genetic variations (i.e., SNPs) to build a classifier model for prediction. In this study, SNPs were used as markers for hypertension in patients. We utilized the OpenSNP dataset, which includes 19,697 SNPs from 2,052 samples. Extreme gradient boosting (XGBoost) is an ensemble machine learning method employed here for feature selection, which incrementally adjusts weights in a series of steps.
Results:
The experimental results identified 292 SNPs that exhibited high performance, with an F1-score of 98.55%, precision of 98.73%, recall of 98.38%, and overall accuracy of 98%. This study provides compelling evidence that the XGBoost feature selection method outperforms other representative feature selection methods, such as genetic algorithms, analysis of variance, chi-square, and principal component analysis, in predicting hypertension risk, demonstrating its effectiveness.
Conclusions
We developed a model for predicting hypertension using the SNPs dataset. The high dimensionality of SNP data was effectively managed to identify significant features as biomarkers using the XGBoost feature selection method. The results indicate high performance in predicting the risk of hypertension.
6.Pseudomonas aeruginosa in Chronic Suppurative Otitis Media
Artono ARTONO ; Nyilo PURNAMI ; Edi HANDOKO ; Agung Dwi Wahyu WIDODO ; Juniastuti JUNIASTUTI
Infection and Chemotherapy 2025;57(1):63-71
Background:
Chronic suppurative otitis media (CSOM) has caused many hearing disorder cases in developing countries.Inappropriate antibiotic use resulted in a shift of bacterial resistance. The biofilm-forming bacteria, like Pseudomonas aeruginosa, was a common germ detected in CSOM that contributed to a poor prognosis. This study aimed to investigate the bacterial pattern from samples taken from CSOM patients regarding its antibiotic susceptibility and the antibiofilm activity of acetic acid against P. aeruginosa.
Materials and Methods:
Sterile swabs of forty-five patients with CSOM were collected, followed by isolation of bacterial pathogens, identification, and evaluation of antibiotic sensitivity using modified Kirby Bauer disc diffusion protocol. In vitro testing was done by adding acetic acid to P. aeruginosa culture to gauge the minimum concentration of biofilm inhibition and eradication. They were conducted using the microtiter plate assay method and quantified with an ELISA reader. The data were analyzed statistically using One-Way ANOVA and Tukey Honestly Significant Difference post hoc test.
Results:
The samples obtained from 31 of 45 CSOM patients showed positive microbial growth; 26 (57.78%) had a monomicrobial pattern, and 5 (11.11%) had a polymicrobial pattern. The researcher ascertained that 24 isolates, representing 66.67%, were gram-negative bacteria, with P.-aeruginosa identified as the predominant species.P. aeruginosa isolates were sensitive to several antibiotics, including meropenem, amikacin, piperacillin-tazobactam, ceftazidime, and cefoperazone-sulbactam with a rate of 93.33%. The minimum concentration of acetic acid required to qualify as the minimum biofilm inhibitory concentration (MBIC) was determined to be 0.16%, yielding an inhibition rate of 26.79%. A concentration of 0.31% was identified as the minimum biofilm eradication concentration (MBEC), achieving an eradication rate of 77.27%.
Conclusion
P. aeruginosa, the most common bacteria found in CSOM samples, was sensitive to imipenem, amikacin, piperacillin-tazobactam, ceftazidime, and cefoperazone-sulbactam. Acetic acid suppresses P. aeruginosa bacterial biofilm formation at MBIC of 0.16% and MBEC of 0.31%.
7.Diagnosis and Management of Hepatic Hydrothorax
Amie VIDYANI ; Citra Indriani SIBARANI ; Budi WIDODO ; Herry PURBAYU ; Husin THAMRIN ; Muhammad MIFTAHUSSURUR ; Poernomo Boedi SETIAWAN ; Titong SUGIHARTONO ; Ulfa KHOLILI ; Ummi MAIMUNAH
The Korean Journal of Gastroenterology 2024;83(2):45-53
Hepatic hydrothorax is a pleural effusion (typically ≥500 mL) that develops in patients with cirrhosis and/or portal hypertension in the absence of other causes. In most cases, hepatic hydrothorax is seen in patients with ascites. However, ascites is not always found at diagnosis and is not clinically detected in 20% of patients with hepatic hydrothorax. Some patients have no symptoms and incidental findings on radiologic examination lead to the diagnosis of the condition. In the majority of cases, the patients present with symptoms such as dyspnea at rest, cough, nausea, and pleuritic chest pain. The diagnosis of hepatic hydrothorax is based on clinical manifestations, radiological features, and thoracocentesis to exclude other etiologies such as infection (parapneumonic effusion, tuberculosis), malignancy (lymphoma, adenocarcinoma) and chylothorax. The management strategy involves a stepwise approach of one or more of the following: Reducing ascitic fluid production, preventing fluid transfer to the pleural space, fluid drainage from the pleural cavity, pleurodesis (obliteration of the pleural cavity), and liver transplantation. The complications of hepatic hydrothorax are associated with significant morbidity and mortality. The complication that causes the highest morbidity and mortality is spontaneous bacterial empyema (also called spontaneous bacterial pleuritis).
8.Effect of Spirulina on Cochlea Histopathological Changes in Wistar Rats Induced by Kanamicin
Prima Erlangga Harinto ; Dian Ayu Ruspita ; Dwi Marliyawati ; Pujo Widodo ; Zulfikar Naftali
Malaysian Journal of Medicine and Health Sciences 2023;19(No.2):69-75
Introduction: Streptomycin and kanamycin are aminoglycosides that are toxic to the cochlea vestibular system, can
causing hearing loss. This antibiotic is used for the treatment of tuberculosis and its ototoxicity occurs in 20% of
tuberculosis patients. Spirulina is a cyanobacterial species that is used as a dietary supplement and contains phycocyanin compounds that function as antioxidants and anti-inflammatory. The aim of this study was to determine the
effect of spirulina on histopathological changes in the cochlea in Wistar rats after kanamycin induction. Methods:
this study is a form of posttest-only controlled group design research with a sample of 24 wistar rats divided into
4 groups, namely negative control group, positive control group, treatment group 1 and treatment group 2. Observations of the study took place in November-December 2021. Histopathological measurements in hair cells, macrophages and cochlear vasculature. The analysis used non-parametric Kruskal-Wallis and post-hoc Mann-Whitney
tests. Results: There were more hair cell damage, macrophage cell count, and significant vascular dilatation in the
kanamycin group than in the without kanamycin group with the value p=0.001. There was significantly less number
of hair cell damage in the kanamycin group with spirulina at a dose of 1000 mg than in the kanamycin group with
spirulina at a dose of 400 mg p=0.045. Conclusion: There was a significant effect on the administration of spirulina
on histopathological changes in the cochlea of rats.
9.Frey’s Procedure for Chronic Pancreatitis in an Adolescent with Recurrent Upper Gastrointestinal Bleeding: A Case Report
Adianto NUGROHO ; Ariani Dewi WIDODO ; Indah JAMTANI ; Asri Dwi RACHMAWATI
Korean Journal of Pancreas and Biliary Tract 2022;27(4):164-167
Chronic fibroinflammatory pancreatitis causes irreparable damage to the pancreatic parenchyma. This frequently results in food restrictions, painkiller addiction, and serious quality of life impairment in children. We observed a 13-year-old girl who had previously been hospitalized multiple times and had undergone acute pancreatitis arrived with excruciating abdominal pain and recurrent hematemesis. A chronic intra-pseudocyst bleed and an ampulla of Vater hemorrhage were discovered during an upper gastrointestinal endoscopy. A 3×2×1 cm pancreatic head pseudocyst was discovered on the magnetic resonance cholangiopancreatography, however the computed tomography scan revealed a pancreatic head pseudocyst, pancreatic duct stones, and substantial peripancreatic inflammation. The multidisciplinary team determined that Frey’s surgery was the most efficient method to minimize her suffering. No complications occurred during the healing phase following surgery, and two years later, neither recurrence bleeding nor abdominal pain appeared. In summary, Frey’s method is a secure and efficient intervention when applied by a multidisciplinary team.


Result Analysis
Print
Save
E-mail