Pattern Recognition Lab

29/05/2026

Glad to be part of a delegation from FAU, Evangelische Hochschule Nürnberg and Siemens Healthineers today – and what a day it turned out to be! We spent the morning at the West China Hospital Museum of History, stepping into a space that holds more than a century of medical heritage. Walking through the exhibition, I was struck by how the stories of early physicians, the evolution of surgical tools, and the cultural context of health care in China all weave together.

Seeing the original instruments and old patient records reminded me why I love bridging academia, industry and history. It’s one thing to read about breakthroughs in a paper; it’s another to stand in a room where those breakthroughs first took shape. The museum does a fantastic job of making that past feel alive, and I left with a fresh appreciation for the global journey of medicine.

A big thank you to my colleagues at FAU and Evangelische Hochschule Nürnberg for organizing the trip, and to Siemens Healthineers for making the visit possible. It’s inspiring to share these moments with fellow researchers, educators and industry partners who all care about where we’ve come from and where we’re headed.

I’m already thinking about how the insights from today could spark new ideas for our own projects back home. If you ever get the chance to visit the West China Hospital Museum of History, I highly recommend it – it’s a reminder that every breakthrough stands on the shoulders of countless stories that came before.

29/05/2026

Reconstruct Anything Model: a Lightweight General Model for Computational Imaging

Imagine trying to recover a perfect photograph from a blurred snapshot taken in low light, or trying to reconstruct the precise internal structure of the human body from sparse, noisy data collected by an MRI machine. These are not simple digital cleanups; they are profound challenges that sit at the heart of **computational imaging**. At its core, this field grapples with what mathematicians call "inverse problems."

The Mystery of Imperfect Pictures and the Dawn of Universal Imaging

29/05/2026

Great pleasure to visit West China Hospital’s Orthopedics Department together with colleagues from Siemens Healthineers. It was inspiring to walk the halls where cutting-edge orthopaedic care meets real-world patient needs.

We spent the day sharing results, swapping ideas and brainstorming how to bring medical AI straight to the bedside. Seeing the enthusiasm of the clinicians there reminded me why we’re all in this – to make diagnostics smarter, treatments more precise and everyday care better for every patient.

I was especially pleased that this year FAU was joined by specialists from the Lutheran University of Applied Sciences Nuremberg. Their fresh perspectives added a valuable layer to the conversation and highlighted how interdisciplinary collaboration can accelerate progress.

A big thank you to the West China Hospital team for hosting us and to the Siemens Healthineers partners for their technical expertise. The exchange of knowledge felt like a true step forward for medical AI, and I’m excited to keep the momentum going.

Looking ahead, I’m hopeful that the ideas we sparked together will translate into tangible improvements in patient outcomes. Stay tuned for more updates as we turn these discussions into action.

28/05/2026

BAIOSPHERE Medical 2026 is seeking motivated student volunteers who reside in or near Erlangen to support the on-site operations of the upcoming workshop. The two-day event, scheduled for 8–9 June 2026 in Erlangen, will convene clinicians, imaging scientists, AI researchers, computer scientists, engineers and translational investigators to discuss developments at the intersection of medicine, imaging and artificial intelligence.

Volunteers will assist with tasks that ensure a welcoming and smoothly run conference for participants. In exchange, volunteers receive complimentary admission, a BAIOSPHERE Medical event t-shirt and unrestricted access to all scheduled activities. Because the role requires on-site presence before and during the workshop, applicants should be based in Erlangen or the surrounding area.

Students with an interest in artificial intelligence, medical imaging, healthcare innovation or the logistics of scientific meetings are encouraged to apply. Registration is handled through the provided form:

https://docs.google.com/forms/d/e/1FAIpQLSfPBx9_mI_TRPvN4fzhSb3TbOCEX-IVjYznIEjR03RdVquHwQ/viewform?usp=send_form

28/05/2026

Unfortunately I already had to leave India because a workshop in Chengdu, China called me away. It was a bittersweet goodbye to the amazing people I met in India, but the journey continues.

Now I’m back in Chengdu, re-uniting with the other members of FAU. We’ve already gathered around a table of delicious Chinese food, sharing stories, laughs, and a few extra dumplings. It feels great to catch up in person after all the virtual meetings.

The next few days are packed with meetings, talks, and joint research sessions with West China Hospital. I’m excited to dive into the collaborations we’ve been planning and see where our combined expertise can take us. There’s a real buzz in the air – the kind of energy that only comes when brilliant minds come together in the same room.

I also want to take a moment to thank our partners BayChina and Siemens Healthineers. Their support has been essential for our joint endeavors, and I’m grateful for the resources and enthusiasm they bring to the table.

Looking forward to the conversations ahead, the ideas that will spark, and the progress we’ll make together. Stay tuned for updates from Chengdu – I’ll be sharing more about the talks, the research, and maybe a few more food photos!

28/05/2026

Just Ask for a Table: A Thirty-Token User Prompt Defeats Sponsored Recommendations in Twelve LLMs

The rise of Large Language Models (LLMs) has ushered in an era of incredibly powerful conversational tools, transforming how we seek information, book travel, and solve problems. These sophisticated assistants are becoming integrated into the very fabric of our digital lives. Yet, beneath the veneer of helpful, objective guidance lies a growing commercial reality. AI chatbot developers are increasingly weaving advertising into these helpful interfaces. The question, therefore, is not just whether these models *can* be influenced by commercial interests, but whether they are *discreetly* doing so, nudging users toward paid, sponsored options without ever revealing that bias.

The Shadowy Commerce of AI Chatbots

27/05/2026

Today I had the chance to visit Bangalore’s Center for Brain Research at the Indian Institute of Science (IISc). Walking through their labs was like stepping into the future of neuroscience – the energy there is palpable and the work truly inspiring.

The team walked us through their massive brain-health cohort study. It’s a truly interdisciplinary effort: they’re collecting multimodal imaging, speech recordings, motion capture, and even omics data from thousands of participants. Seeing how they integrate all those data streams gave me a concrete sense of how AI can be harnessed to unlock new insights into brain health. The scale and rigor of the study are impressive, and the foundational work they’re doing feels especially timely as we all grapple with the possibilities – and responsibilities – of AI in medicine.

What struck me most was the collaborative spirit. Researchers from different backgrounds were chatting about everything from data pipelines to the ethical dimensions of their work. It reminded me why I love this field: it’s not just about the tech, but about people coming together to solve big, real-world problems.

Hope to be back soon for more deep dives, more conversations, and more inspiration. If you’re curious about the intersection of AI and brain research, keep an eye on what’s happening at IISc – they’re doing groundbreaking stuff that will shape the next decade of discovery.

27/05/2026

I’m thrilled to share a new milestone for our team at FAU. Today our pattern recognition lab signed an MOU with the School of Computer Science and Engineering at RV University. It’s a concrete step toward deepening our ties with India and opens the door to exciting collaborations.

The partnership means we’ll be working closely with colleagues in Bangalore, sharing expertise, joint projects, and student exchanges. I’m looking forward to the research opportunities this will create and to building lasting relationships across continents.

A big thank-you to everyone who helped make this happen – from our lab members to the leadership at RV University. Here’s to many fruitful endeavors ahead!

27/05/2026

Fantastic research that I could learn to know here at the two-day workshop “Machine Learning and Deep Learning for Multimodal Neurological Disorder Detection” organized by RV University, Bangalore. It was inspiring to see participants from FAU, MAHE, PES University, Nitte and NIMHANS all gathered around the same challenge.

The talks and demos showed how far we’ve come in applying AI to complex neurological data, and the energy in the room was contagious. I’m really excited about what has been done and the ideas that are already taking shape.

Beyond the great science, I’m hopeful that this gathering will spark mutual, long-term collaborations. There’s a lot we can achieve together—sharing data, co-authoring papers, and building tools that make a real difference for patients.

Thank you to the organizers, the speakers, and everyone who contributed their time and expertise. Looking forward to staying in touch and turning today’s conversations into tomorrow’s breakthroughs!

27/05/2026

High-Fidelity Postmyocardial Infarction Ventricular Tachycardia Simulation for Intraprocedure Ablation Guidance

The human heart, a marvel of biological engineering, is not immune to failure. After a heart attack—a myocardial infarction—the tissue that has been damaged forms scar tissue. This scarred area is not electrically inert; instead, it often becomes a chaotic playground for electrical signals, leading to dangerous abnormal rhythms known as ventricular tachycardia (VT). Imagine the heart’s engine starting to rev wildly and inefficiently, threatening to collapse into ventricular fibrillation—a fatal, chaotic quiver.

The human heart, a marvel of biological engineering, is not immune to failure.

26/05/2026

I just got back from a wonderful project meeting at the NIMHANS MR PET Center. It was great to sit down with Dr. Rose and the Siemens Healthineers team and dive into what we can achieve together.

The conversation was energising – we walked through a handful of ideas, mapped out some realistic next steps, and quickly saw where our expertise overlaps. I left the room feeling confident that we’ll be able to launch several joint projects in the near future.

What struck me most was the shared enthusiasm for pushing the boundaries of MR-PET imaging. Dr. Rose brought deep clinical insight, while the Siemens Healthineers engineers showed impressive technical know-how. It felt like the perfect blend of science and technology, and I’m excited to see how our collaboration will translate into real-world impact for patients.

A big thank-you to everyone who made the meeting possible – the staff at NIMHANS, Dr. Rose for her leadership, and the Siemens Healthineers team for their openness and vision. I’m looking forward to the next steps, drafting proposals, and getting our teams rolling.

Stay tuned for updates as we move forward. It’s an exciting time to be working at the intersection of AI, imaging, and healthcare, and I’m grateful to be part of a community that’s so eager to innovate together.