Internship: Exploring AI Algorithm Implementation on FPGA

Internship: Exploring AI Algorithm Implementation on FPGA

Start date: September or October 2025

Duration: Minimum 5-6 months, longer is possible and preferred

Type: Thesis and non-thesis internship is possible

Modern medical imaging systems depend on real-time image processing to support physicians during minimally invasive procedures. These systems combine specialized hardware (e.g., FPGAs), embedded software, and PC-based applications to deliver high-performance, low-latency results.

As expectations for image quality, responsiveness and guidance increase, pre-trained AI models are becoming integral in areas like noise reduction, enhancement, and pattern recognition. While CPUs and GPUs are widely used for AI, FPGAs offer unique advantages: predictable timing, low latency, and hardware-level parallelism which is critical for real-time medical use cases.

To make AI deployment on FPGAs more accessible, new tools and frameworks allow engineers to implement inference models without writing low-level hardware description code. This opens opportunities to explore how existing trained models can be mapped to FPGA hardware using various flows.

Problem description:
Most AI models are developed and trained in high-level environments like Python and C++. Traditional FPGA development, using VHDL or Verilog, poses a barrier to rapidly deploying such models.

Emerging toolchains now support flows where trained models can be translated into synthesizable logic using high-level languages or model conversion frameworks. This assignment challenges the student to explore the state of the art in AI-on-FPGA inference, evaluate implementation options, and build a working prototype using a simple AI algorithm.

Your role

1. Research & Landscape Analysis

   • Investigate the AI inference capabilities of the FPGA platforms used in medical imaging systems, focusing on hardware features such as DSP blocks, memory hierarchies, and parallel processing units.
   • Conduct a survey of available toolchains and frameworks (e.g., Vitis AI, FINN, hls4ml, Brevitas, ONNX, Deep Learning IP cores) that enable mapping of pre-trained AI models to FPGA hardware.
   • Summarize state-of-the-art trends in AI-on-FPGA deployment, with an emphasis on their relevance to real-time, safety-critical medical imaging applications.

2. Technology Exploration

   • Perform a comparative study of FPGA development flows for AI inference:
   • • Low-level RTL (VHDL/Verilog)
     • High-Level Synthesis (C++/SystemC)
     • Model-to-hardware compilers and frameworks (e.g., FINN, hls4ml).
   • Evaluate which types of AI models (e.g., quantized CNNs, MLPs, statistical algorithms) are most feasible for FPGA deployment.
   • Assess common optimization techniques (quantization, pruning, fixed-point arithmetic) and their impact on resource usage, latency, and accuracy.

3. Prototype Implementation

   • Select a representative AI algorithm or lightweight statistical method relevant to medical image enhancement (e.g., noise reduction filter, basic classifier, or histogram-based analysis).
   • Implement the algorithm, synthesize it to FPGA, where relevant, experiment with modern AI-to-FPGA frameworks.
   • Optimize the design to balance accuracy, speed, and hardware efficiency.

4. Demonstration & Evaluation

   • Integrate the implemented algorithm into a test pipeline with real or representative image data.
   • Demonstrate functionality on FPGA hardware.
   • Present a comprehensive evaluation covering performance, latency, resource usage, and feasibility for use in real-time medical imaging workflows.
   • Provide recommendations for future work, highlighting opportunities for scaling to more complex AI models or clinical use cases.

Deliverables

• Research Report – Summary of FPGA capabilities, frameworks, and trends.
• Technology Evaluation – Comparative study of development flows, models, and optimizations.
• Prototype Package – Source code, synthesizable implementation, and reproducible build instructions.
• Demo & Results – Working demonstration with image data and performance evaluation.
• Final Presentation – Findings, results, and recommendations for future exploration.

You’re the right fit if

• Pursuing a Master degree in Electrical Engineering, Computer Engineering, or Artificial Intelligence.
• Experience with FPGA design (VHDL/Verilog) or High-Level Synthesis (C++/SystemC)
• Familiarity with AI/ML fundamentals (e.g., CNNs, quantization, pruning)
• Interest in FPGA AI toolchains (Vitis AI, FINN, Brevitas, hls4ml)
• Basic knowledge of Python for handling models and frameworks
• Research-oriented mindset with curiosity for new technologies
• Ability to balance theory with hands-on prototyping
• Strong communication skills for presenting results

In return, we offer you

A non-stop growth opportunity within a global multinational company. In addition, we also want to remunerate you well for the opportunity. This includes:

• Monthly full-time allowance between €500 - €700 gross, depending on your education level and if you need to relocate to The Netherlands. Part-time internships receive the allowance pro-rata.
• Housing compensation. Should you need to relocate to near our office, we’ll offer you a monthly allowance of €300 net if location is in the vicinity of Amsterdam, €255 net for all other Philips locations. To be eligible, we’ll just ask for the rental contract from your landlord, and your usual home-work travel distance needs to be more than 50 km or 1.5 hours of travel time (for a one way trip).
• Travel compensation, if you are not eligible for a free public transport card; you will receive max. €192 net as a monthly allowance.
• Paid holidays per internship term.

How we work at Philips
We believe that we are better together than apart. For our office-based teams, this means working in-person at least 3 days per week.

About Philips
We are a health technology company. We built our entire company around the belief that every human matters, and we won't stop until everybody everywhere has access to the quality healthcare that we all deserve. Do the work of your life to help the lives of others.