Dennis Bontempi, Ph.D.

Doing fun imaging genetics research @dbc_unil 🫀🧬
Former Research Scholar @aim_harvard - PhD @MaastrichtU

I love coding, reading, science & a bunch of sports!

[The Lancet Digital Health @LancetDigitalH]

NEW Research: FaceAge, a #DeepLearning system to estimate biological age from face photographs to improve prognostication: a model development and validation study. @HugoAerts @denbonte @Dr_RayMak

Read it here: https://t.co/fAAWybFCG5

[Dennis Bontempi, Ph.D. @denbonte]

Why should we care? 🧠✨

There is still much to learn about #cardiovascularhealth and looking at the combination of both cLD and cIMT may be better than either one alone (... more on this coming soon 👀)

#research #genetics #ultrasound #AI #UKBiobank

[Dennis Bontempi, Ph.D. @denbonte]

New #preprint alert 🚀 (hopefully the first of many at @dbc_unil @unil 😎)

https://t.co/QblI3ZlBRX

... Did you know arteries can expand to compensate for bad cardiovascular health? A thread! 🫀🧬🧵

[Dennis Bontempi, Ph.D. @denbonte]

👉 We also investigated the ratio of cIMT to cLD, looking for a new perspective on the relationship between arterial wall thickening and lumen diameter that might lead us to a deeper understanding of arterial health!

[Dennis Bontempi, Ph.D. @denbonte]

Thrilled to have helped in this😎

Congrats to the team, and many thanks to @FabianIsensee1 & @JakobWasserthal (and collaborators!) for the #OpenSoftware that made this possible! 🚀

[Ray Mak, MD @Dr_RayMak]

🔬Just out in @LancetDigitalH , our work on AI for #patientsafety in #RadOnc

https://t.co/BxrNShCk5A

Tldr
🎯 Problem: Wrong level spine treatment in medicine is a significant safety risk (counting/labeling correct vertebra is error prone, and some ppl have variant # of vertebrae).

🤖 Solution: We developed and prospectively implemented a #AI quality assurance tool into a #radiotherapy clinical workflow to check for wrong level spine targeting as an assistive technology (think lane departure assist in cars) and caught rare errors before they reached the patient, improving patient safety.

#MedTwitter #DigitalHealth #AI #Oncology @astro_org @OncoAlert @HugoAerts

[Dennis Bontempi, Ph.D. @denbonte]

Some great research coming out of @dbc_unil 🙌

... this one made the Nature cover as well!

[nature @Nature]

By analysing the ancient genomes of individuals from Rapa Nui, researchers have overturned a contentious theory that the remote Pacific island experienced a self-inflicted population collapse before European colonization https://t.co/dCICVqCUxJ

[Dennis Bontempi, Ph.D. @denbonte]

@alfcnz @dkpublishing @WHOOP Great! Just stick with what you like, I don’t think they are much different!

Apple is also introducing Vitals now which is very close to what Whoop does (although with a much less precise HRV measurement probably, due to sampling frequency) 🙃

[Dennis Bontempi, Ph.D. @denbonte]

@alfcnz @dkpublishing @WHOOP … just to make sure you have a grasp on how your body reacts to changes in exercise, nutrition, and so on. Bonus: Whoop does some cool research as well.

And of course, besides wearables, a nice app to track lifts and progress - StrengthLog.

You have this 💪🔥

[Dennis Bontempi, Ph.D. @denbonte]

@alfcnz @dkpublishing Last but not least (sorry 😂), it wouldn’t be a great experimental setup without something to measure changes with.

If you get seriously into fitness, sleep and recovery are crucial - so I would consider investing in something like @WHOOP (I love mine!) to track progress.

[Dennis Bontempi, Ph.D. @denbonte]

@alfcnz Also, I would advise you to find a good nutritionist and invest some time in that aspect of body recomposition. Always worth the time and money (and I know you enjoy cooking, so that's also good fun).

For the rest, be patient and trust the process. Excited to see the progress!💪

[Dennis Bontempi, Ph.D. @denbonte]

@alfcnz ... just to name a few. I am confident you will be able to navigate that YT community pretty well.

They have so much free material, beautifully laid out, that I think you will struggle to find something more helpful.

[Suraj Pai @BSurajPai]

Contrastive learning of representations can help discover quantitative imaging biomarkers for oncology!
📝 https://t.co/a6GcB82ZmQ

☑️ Performant in low data
☑️ Resource-efficient
☑️ Robust to inter-reader and acquisition diff.

@NatMachIntell @aim_harvard @BrighamRadOnc
🧵1/6

[Dennis Bontempi, Ph.D. @denbonte]

If you're still unsure about who to vote for the MSB Educational Challenge - I've just asked ChatGPT.

... One does not simply ignore ChatGPT.

#AcademicTwitter #MICCAI2023 #Medical #Imaging #MRI #Python #DataScience

[MICCAI Students @MiccaiStudents]

🌟 Finalist Spotlight: Anna Zapaishchykova & @BenjaminKannMD
🎥 "Preprocessing MRI in Python"

Dive into this insightful tutorial and explore innovative approaches to MRI preprocessing.

👉 Watch now: https://t.co/4WJQvDRyh6