Introduction to MacsLAB
Our main goal is to develop AI applications for real-world medical use,
including disease diagnosis, lesion detection, and image quality enhancement
through medical data analysis and processing.
🎯 Research Goals
MacsLAB goes beyond simple model development —
we aim for the practical deployment of AI systems that can be applied in clinical environments.
To ensure AI supports medical professionals’ decision-making,
we study AI systems that balance accuracy, explainability, and reliability.
🧬 Main Research Areas
1️⃣ Medical Image Analysis
- DICOM-based image processing and standardization
- Automatic detection models for fractures, tumors, and abnormal signals
- Segmentation and classification of CT / MRI / X-ray images
- Quantitative imaging biomarker computation
2️⃣ Deep Learning–based Computer Vision for Medicine
- Lesion detection using CNN and Vision Transformer (ViT)
- Application of self-supervised and few-shot learning methods
- 3D medical volume reconstruction and registration
- Research on AI visualization (Grad-CAM, SHAP, Saliency Map)
3️⃣ Clinical Application and Explainability of AI Models
In the medical field, AI misjudgment can endanger lives.
Therefore, ensuring interpretability, reproducibility, and safety is crucial.
MacsLAB builds stable and transparent AI deployment and visualization systems,
considering the unique characteristics and regulations of medical data.
🧩 Research Example
Lesion Detection Model in Medical Imaging
Example Code
import torch
import torch.nn as nn
import torchvision.models as models
class AFFDetector(nn.Module):
def __init__(self, num_classes=2):
super().__init__()
self.backbone = models.resnet50(pretrained=True)
self.fc = nn.Linear(1000, num_classes)
def forward(self, x):
x = self.backbone(x)
return self.fc(x)
model = AFFDetector()
print(model(torch.randn(1, 3, 224, 224)).shape)
