Are your fingerprints really unique?

Summary

Highlights

• A surprising misconception busted: Each finger has a unique fingerprint, even on the same hand! 👋
• AI is making waves by identifying fingerprints beyond traditional minutiae analysis. 📈
• The basic fingerprint patterns consist of arches, whirls, and loops. 🌊
• AI detected subtler patterns in fingerprints, showing promise in forensic advancements. 🕵️
• Fingerprints form in the womb and remain constant from birth, not changing over time. 📅
• Even AI, when tested on public data, matched fingerprints with 80% accuracy using new criteria. 🎯
• AI’s potential is not just in forensics but also in medical diagnostics for identifying diseases. 🩺

Key Takeaways

• Every finger has a unique fingerprint, even if they belong to the same person! 🤔
• AI technology is changing how we view fingerprint identification. 🔍
• Different patterns like arch, whirl, and loop classify fingerprints, but real uniqueness lies in minutiae. 👀
• Even identical twins don't share the same fingerprints! 🤯
• The minutiae of fingerprints are formed in the womb and remain unchanged throughout our lifetime. 🚼
• AI can sometimes outperform human analysts in recognizing fingerprints from the same person. 🤖
• The discovery of AI’s pattern recognition could revolutionize both forensic and medical fields. 💡

Overview

Chapters

• 00:00 - 00:30: Introduction to Fingerprint Matching The chapter titled 'Introduction to Fingerprint Matching' involves an interactive forensic exercise where a participant, Kim, is tasked with matching fingerprints. Two of the fingerprints collected by the narrator are from the same finger, and the challenge is to identify these matching prints. Kim, a video producer with a background in health, science, history, and culture, participates in this engaging demonstration as part of the 'explainer Club' series, which features video producers explaining topics to their coworkers.
• 00:30 - 01:30: Fingerprint Patterns and AI Tool The chapter delves into fingerprint patterns and how AI tools can assist in identifying them. Initially, there is a discussion around making guesses based on visual observations. The narrator starts by hypothesizing that two fingerprints, labeled A and B, are the same due to a 'little swirl' pattern and distinct dissimilarity in the other samples. Later, they express uncertainty over their guess and reflect on the outcome where the correct matching was actually A and C. The exploration touches upon left-hand fingerprints, noting a mistake in original assumptions. Overall, the chapter illustrates the complexity and potential inaccuracies in manually evaluating fingerprint patterns without AI assistance.
• 01:30 - 02:30: Fingerprint Cards and Patterns The chapter explores the uniqueness of fingerprints, highlighting the fact that each finger has a different print even on the same person. The discussion includes an interesting point about police procedures for fingerprinting, where all 10 fingers are taken for a complete reference. This raises questions about the complexity of identifying individuals if each fingerprint is different.
• 02:30 - 04:00: Minutiae and Fingerprint Uniqueness In this chapter, the uniqueness of fingerprints and recent technological advancements in fingerprint matching using AI are discussed. The conversation highlights traditional fingerprint matching methods used by professionals, like the FBI, and how challenging it can be to capture accurate fingerprints without proper training. A new AI tool that claims to improve fingerprint matching efficiency and accuracy is introduced.
• 04:00 - 06:00: AI and Fingerprint Analysis The chapter discusses the FBI's fingerprint archive, which contains around 165 million fingerprint records. These records originate from fingerprints taken at the time of arrest in the US, irrespective of the final outcome of guilt. The chapter highlights the three primary patterns used in fingerprint classification: the arch, whirl, and loop. There is a mention of the need for prior information before quizzes, as noted by a person named Coleman.
• 06:00 - 08:00: AI in Medicine and Future Outlook The chapter discusses AI in medicine, emphasizing its potential and future outlook. It explains the categorization of fingerprint patterns as an analogy, describing the different patterns like arch, loop, and whirl. The loop is noted as the most common pattern, while whirl is less common but not rare. The explanation of fingerprint patterns is used to metaphorically discuss the variety and complexity of cases AI might handle in medical diagnostics and treatments, highlighting the potential for AI to identify unique patterns akin to fingerprints in medical data.
• 08:00 - 10:00: Conclusion and Reflections The conclusion and reflections chapter discusses the intricacies of fingerprint analysis. It emphasizes that general fingerprint patterns are more about classification, whereas the actual identification relies on the minutiae. Key features in fingerprint analysis include ridge endings, where a ridge stops, and bifurcations, where a ridge splits into two. The chapter also mentions the rarity of arches and introduces the concept of an 'island,' a small dot formed when a ridge splits but then quickly stops.

Are your fingerprints really unique? Transcription

• 00:00 - 00:30 so Kim I'm might you try a little amateur forensic work here I've collected fingerprints from a few of our colleagues and what you see here is four fingerprints I took two of which are from the exact same finger so I want to see if you can match which two prints are from the same finger this is explainer Club where video producers like me sit down and explain something to one of our co-workers and today that is Kim I'm a video producer and I work on primarily health and science content and I do history culture and sometimes
• 00:30 - 01:00 science videos so I think A and B that's my first guess and that's just because I see this like little swirl coming up and these two just look so different cuz you have like this little hump down here I think I still want to say A and B are the same okay I feel like I completely blew that I feel like the okay was a little like you are the answer was a and C oh shoot they're both from the left some so what
• 01:00 - 01:30 is interesting to me about this though is all of these fingerprints are from the same person oh I think we all know every person has their own fingerprint but I didn't know before I started working on this that each individual finger on your hands is different from each other did you know that like they're completely different police practice when you get fingerprinted they'll take all 10 that makes sense yeah but that makes it seem like it's so much harder to actually identify if a fingerprint belongs to a person if every single one is different right without this full reference you wouldn't be able
• 01:30 - 02:00 to match one fingerprint to another and say this came from the same hand or even the same person but there is this new AI tool that says it can do it that's what we're going to talk about today okay cool yeah cool we're going to probably run the logo there logo okay show you something else this is a copy of the actual fingerprinting card that the FBI uses this one is kind of a practice for me I learned while making this video that taking someone's fingerprint if you're not training to do that is actually kind of hard did it work no I
• 02:00 - 02:30 think I it up the FBI's fingerprint archive currently houses around 165 million fingerprint records and they start on these cards so if you're arrested for a crime in the US you will get your fingerprints taken by the police and they're stored on file even if you're not proven guilty so if you flip this over these are the three main patterns used to classify fingerprints the arch whirl and loop okay I feel that you should have given me this before the quiz Coleman I feel I wasn't
• 02:30 - 03:00 Fair okay that is fair but no this is cool so the ridges in an arch pattern come from one side of the fingertip and then exit out the other side then a whirl spirals into a center point and a loop comes from one side of the finger and then back out the same side this little bottom part's like an arch right and this is a loop cuz this is coming back this way yeah exactly not a lot of whirls going on is whirl more rare it's kind of in the middle about 60% of people have loops that's the most common whirl is considered like around 35% and
• 03:00 - 03:30 then arches are actually very rare oh interesting these General categories are really just for classification actually analyzing and matching an individual fingerprint is in the minutia the little brakes and splits along each individual Ridge not the overall pattern and the most common markings are when you follow a ridge and it ends that's called a ridge ending and then if you follow along a ridge and it splits into two that's called a bifurcation there's an island which is like that little dot when a ridge splits but then quickly
• 03:30 - 04:00 rejoins is called a lake and then that's a bridge connecting two ridges so analyst will basically just go ridge by Ridge and Mark these details and that is where uniqueness and fingerprints really comes from not even identical twins who have matching DNA have matching fingerprints when you get to the minutia so the minutia is how analysts have been matching fingerprints for more than a century is that coming from outside factors like it's not genetic anymore like if I burn my finger or something like that is that where most of mini is from that's a great question it doesn't
• 04:00 - 04:30 come from your lifetime I'm just thinking of like Men In Black where Will Smith has to like grab the thing and it just burns his fingerprints off one you encounter that would be hard to actually do it's very hard to get rid of your fingerprints and they don't change over your lifetime they're actually tiny Ridges of tissue under your skin oh they're called friction ridges even if you burned off your fingertips your palms your other joints they're all equally unique cuz they have the same friction ridges so I want to show you something on my computer by the time you're born your fingerprints are
• 04:30 - 05:00 already set so they form in the womb their pattern begins with some genetic influence from your parents as a shape in the center the fingertip and top of your first knuckle The Ridges grow from there in waves following what's called a turing pattern this is a concept proposed by Alan Turing in the 1950s that helps explain other unique patterns in nature like how a leopard gets its spots how zebra gets its stripes and sandunes that's pretty it's beautiful and so what I want to show you is there some cool simulators online that can kind of give you an idea of how a turning pattern
• 05:00 - 05:30 how they work oh that's so fun oh it looks like a brain that's creepy these waves and and the way they merge and Collide ends up with a unique pattern so I want to ask you about the video that you made about how our voice is like a fingerprint too as humans each of us produces a sound that's about as unique as a fingerprint so the whole point of that video was to just explain the amount of things that go into producing your voice your nasal cavity your tongue your vocal cords have different thicknesses and lengths there's just so many minuscule little factors is that
• 05:30 - 06:00 will just change how you sound and therefore we went with why your voice is like a fingerprint people in the comments were just like actually fingerprints are not unique the thing is is like we can't prove it I think maybe that's what they mean I could see the doubt and I've seen questions online like they didn't take everybody's fingerprint how do they know they're all unique it's a probability issue especially the way these things form under our skin which is truly random and there's just so many data points in here that it was estimated even sort of towards the beginning in the history of fingerprinting the odds of having the exact same pattern when you get down to
• 06:00 - 06:30 the minutia are like 1 in 64 billion and I think that people feel comfortable enough saying that no two fingerprints are alike but about a year ago that might have changed I'm going to play a clip from a phone interview that I did uh recently this is a transcript of what you're about to hear take a listen to this my name is H lipon I'm a roboticist AI researchers here at Columbia and uh have nothing to do with forensics fingerprints anything like that once you understand AI you can apply to taking a
• 06:30 - 07:00 fresh look at some heart problems and this is one of them what he's talking about is this study published in science advances that one of his students was the lead author on that claims to have trained in AI to identify whether a pair of fingerprints came from the same person or not so what they did was feed pairs of fingerprints sometimes from different people sometimes the same person but different fingers into the AI and ask a simple question same person or two different people and so he found the data set which was a set of about 60,000
• 07:00 - 07:30 fingerprints from I think mostly from deceased people is a public data set he set up the AI again all a fancy Ai and had it look at this and lo and behold he was able to do it fairly effectively which led us to ask more questions like how are you doing it what are the features you're using and the AI shows you and it's looking at fingerprints in a completely different way than humans it found similarities to pretty successfully identify whether it was the same person or not how does it do that I
• 07:30 - 08:00 want to show you what it was looking at instead of examining the minutia like a human analyst does the machine focused on the subtle curve shape and different angles of Ridge lines in the center of the fingerprints so what this chart is basically showing is when it examined a raw image the machine was right about 80% of the time when asked the question same person or different people and even when you removed a lot of the detail including the minua and you were left with just this orientation it was still right about 75% of the time so it doesn't need the minutia and in fact fact when they did just the minutia and
• 08:00 - 08:30 removed all other details it was the least effective it was around 60% which is not much better than a random guess that's just to examine if the prints came from the same person or not to match an exact fingerprint to an exact fingerprint you would still need the minutia there's been some push back from experts in the field of forensics about the study who say the paper alone won't change the way we interpret fingerprints and the authors themselves acknowledged that there's more research that needs to be done we took it as far as we can using the public data sets but I'm sure if the FBI wanted to do this they could
• 08:30 - 09:00 do a much better job the a slightly deeper issue is uh Discovery is that we've been looking at fingerprints perhaps we're missing a lot of information because we were overlooking things like curvature and things that AI suggests are actually very meaningful it gives us a new language to look at fingerprints a general optimistic view of AI that I hope to have is how it can teach us to see new things and also just in the field of medical Diagnostics to be able to catch things like tumors or Cancers and like how much we can use
• 09:00 - 09:30 machine learning to improve the lives of people I asked you to do a little bit of homework I was wondering if you had any thoughts just on that General big picture I yeah the interesting thing about AI is that typically when we're reading about it or talking about it or everything there's this like huge future of work thing that comes up where they're just like the robots are going to replace us and I kind of like this other lens that came up when I started looking into you it asked me to look into like the medical field specifically one quote that I saw more more than once
• 09:30 - 10:00 was that AI is not going to replace the doctor it's going to replace the doctor that doesn't use AI the idea there being that doctors will just evolve with this technology and they will start using it to help Advance themselves diagnosing things like cancers how much more quickly can we work in a crisis if we have a machine next to us eliminating the problems these tools raise so many questions I think we're trained to kind of take a really cynical approach with most of them yeah but the idea that it's just we can learn from the machine too I
• 10:00 - 10:30 like that back and forth I like that version of things I like that too that makes me feel better about the machines thanks for watching we wanted to do something different with this video to take our standard explainer video and see what happens when we lay out the explanation in a room with another person for this video that meant doing a lot of my After Effects and design work which is usually the last thing I do early on while I researched and wrote the video and then instead of doing multiple takes in a vo Booth shooting the whole thing in one go and hoping it goes well we've been so focused on the
• 10:30 - 11:00 minutia or minuche minutia I think it's Minier but Kim I should have shown you the basic fingerprint patterns before the quiz I'm sorry about that but the point of this video was the same as always to answer questions about our world and to show you the most interesting things we find in a way that's hopefully fun to watch if you want to support that you can do that at box.com memberships if you're already a member thank you so much you're a big part of the reason that we get to do this at all see you next time