AI & Computer Vision Engineer

I build computer vision and AI systems that turn messy data into reliable decisions.

Master's graduate in Computer Science (AI & Data Science) from the University of Montpellier. At Siemens Healthineers I built a deep-learning pipeline for medical image segmentation — today I do similar work freelance, while staying open to engineering roles in Türkiye or internationally, remote.

View my projects Download CV

M.Sc. AI & Data ScienceUniversity of Montpellier

10 min → 10 secAutomated an MRI analysis workflow at Siemens Healthineers

Freelance AI & SoftwareSince August 2024 — COMANCI

Flagship work

Ovarian follicle segmentation in 3D MRI

Internship at Siemens Healthineers — independently built a computer vision pipeline to automate a medical imaging workflow.

Detected ovarian follicles labeled and outlined on an MRI slice

PythonYOLOv8TensorFlow / PyTorchOpenCVMONAIDICOM / MRI

Problem: Manual segmentation of ovarian follicles on 3D MRI for fertility evaluation is slow and operator-dependent.
Approach: Prepared 3D MRI data (normalization, interpolation, augmentation), evaluated 2D/3D U-Net architectures, then built a YOLOv8-based instance-segmentation pipeline — with post-processing to compute follicle dimensions, volume and location, plus a rule for flagging potential cysts.
Result: ~90% precision, 86% recall on the evaluation dataset. Analysis time dropped from roughly 10 minutes manual to ~10 seconds automated, via a Python interface with CSV and annotated-video export.

See full experience details →

More work

What I've been building recently

Machine Learning

Image Classification — CNN, LIME & Transfer Learning

CNN models trained across multiple datasets, with heatmap visualization, LIME explainability, ResNet50 transfer learning and GAN-based synthetic image generation — paired with an interactive site to explore predictions.

PythonTensorFlowResNet50GAN

View demo → GitHub →

Brain Mapping Research Center · Neuroimaging & ML

Brain Age Prediction from Structural MRI

Iterative pipeline predicting brain age from FastSurfer-derived MRI features, focused on honestly diagnosing and fixing regression-to-the-mean bias rather than chasing an inflated accuracy number.

PythonScikit-learnFastSurferCross-Validation

See case study →

Computer Vision · Freelance

Presence Detection System

Computer vision application differentiating known users from new entrants, built as a freelance client project.

CNNYOLOPython

View all projects →

Recommendation

What people say about my work

"I recommend Ari Cilingiroglu for his outstanding software development project using Artificial Intelligence for structure recognition on medical images. He demonstrated an exceptional level of technical skill throughout the project. Not only did he master complex AI concepts, but he also showed great ingenuity in applying them to solve specific problems in radiology."

Thibaut Eisele — MRI & Syngo Application Manager, Siemens Healthineers · Aug 29, 2024

View the official recommendation