Claim Missing Document
Check
Articles

Found 2 Documents
Search
Journal : Scientechno: Journal of Science and Technology

Analysis of Ferritin Serum in Anemia Pregnant Woman Irianti, Evi
Scientechno: Journal of Science and Technology Vol. 3 No. 3 (2024)
Publisher : Yayasan Pendidikan Islam Daarut Thufulah

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.70177/scientechno.v3i3.1529

Abstract

Anemia remains a common comorbidity complicating over 28% of pregnancy in Indonesia by 2023, Iron deficiency anemia is the most common, and to date, ferritin serum has been used as a diagnostic marker, but has not been routinely examined in pregnant women, thus hindering the accurate diagnosis of iron deficiency anemia in pregnant women. Our study aimed to examine ferritin serum among pregnant women, as well as, to investigate the prevalence of iron deficiency anemia in the Dalu Sepuluh Health Center of Deli Serdang Regency. A total of 60 pregnant women were randomly selected to participate in this cross-sectional study. We found that iron deficiency anemia was most common in pregnant women aged 20 – 35 years old (75%), with second-trimester pregnancy (36.7%), multigravida, and multiparity. Interestingly, 57% of pregnant women were considered malnourished (upper arm circumference of < 23.5 cm) which was found in 36.7% of primigravida and is associated with moderate iron deficiency anemia (71.4%), marked with < 30µg/L ferritin serum (p < 0.05). Cut-off point of serum ferritin was 28.98 ng/dL with a sensitivity of 92% and a specificity of 91.4%, indicating that the body's iron levels are in low condition. This study indicated that lower levels of ferritin serum are associated with anemia in pregnant women. We suggested that ferritin serum be put as a mandatory routine examination in pregnancy.
Pattern Recognition System for Automating Medical Diagnosis Based on Image Data Irianti, Evi; Anis, Nina; Aziz, Saifiullah
Scientechno: Journal of Science and Technology Vol. 4 No. 1 (2025)
Publisher : Yayasan Pendidikan Islam Daarut Thufulah

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.70177/scientechno.v4i1.2126

Abstract

The increasing volume and complexity of medical image data have presented significant challenges for healthcare professionals in delivering timely and accurate diagnoses. Traditional diagnostic processes are often time-consuming and prone to human error, underscoring the need for automated solutions. This study aims to develop a pattern recognition system to automate medical diagnosis using image data, thereby improving diagnostic accuracy and efficiency. A hybrid methodology was employed, combining image preprocessing, feature extraction using convolutional neural networks (CNNs), and classification through deep learning algorithms. The system was trained and validated using publicly available medical image datasets across various disease types. The results demonstrate high diagnostic accuracy, with the system achieving over 92% precision in identifying disease patterns from image inputs. Furthermore, the model exhibited robustness across different imaging modalities, such as X-rays, MRIs, and CT scans. These findings suggest that the proposed pattern recognition system can serve as a reliable support tool for medical practitioners. In conclusion, the integration of image-based pattern recognition in medical diagnostics holds significant promise in enhancing clinical decision-making processes and reducing diagnostic errors.