Ultra-Low-Dose CT Outperforms Chest X-Rays in Paediatric CF
At similar radiation doses, ultra-low-dose CT (ULDCT) achieved superior image quality using AI-enhanced iterative reconstruction than chest x-rays for assessing cystic fibrosis-related lung disease in children. About 88% of healthcare professionals reported greater confidence in diagnosing cystic fibrosis-related lung disease when using CT images rather than x-rays.
METHODOLOGY:
Researchers recruited 75 observers (50 radiographers and 25 radiologists) from 24 countries who assessed the image quality of 10 randomly selected paired ULDCT and chest x-ray images from a cohort of 70 children with cystic fibrosis (age, 3-18 years).
ULDCT scans were acquired at 80 kV, 10 mA with deep learning iterative reconstruction to produce images at lower radiation doses.
Image quality analysis was performed using an image quality survey that was organised into four distinct sections and hosted on DetectED-X.
Observers rated images on the basis of a 0-4 scale rating system for quality and a 5-point Likert scale for diagnostic confidence evaluation.
TAKEAWAY:
Overall, 88% of observers reported higher levels of confidence with ULDCT (P < .05) than with chest x-rays (P < .001) in diagnosing cystic fibrosis-related lung disease, with comparable radiation doses between the two modalities.
The combined visual grading analysis indicated higher image quality for ULDCT images, with radiologists rating ULDCT significantly superior to chest x-rays for lung pathology detection (area under the curve, 0.63; P = .03).
Overall, 69% vs 50% of observers rated the quality of ULDCT images vs chest x-rays as either adequate (score 3) or better than needed (score 4).
No significant differences in anatomic visualisation were observed between ULDCT and chest x-rays (P = .306).
IN PRACTICE:
"ULDCT provides superior IQ [image quality] compared to CXR [chest x-rays] in CwCF [children with cystic fibrosis], while maintaining comparable radiation doses," the authors wrote. "ULDCT also significantly improves diagnostic confidence, highlighting its potential as a viable alternative to CXR," they concluded.
SOURCE:
This study was led by Niamh Moore, University College Cork, Cork, Ireland. It was published online on July 25, 2025, in European Radiology.
LIMITATIONS:
Chest x-rays were used as a comparator instead of conventional dose CT, and exposures were not patient specific for ethical reasons. The recruitment of observers from different research hubs having varying illumination levels affected image quality ratings. Additional limitations included the small sample size and selection bias.
DISCLOSURES:
This study received open access funding provided by the IReL Consortium. One author reported having a relationship with DetectED-X.
This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
CBS News
12 minutes ago
- CBS News
Re-Coding the Future
This content was provided by Acumen Media for Ascendion. This advertiser content was paid for and created by Acumen. Neither CBS News nor CBS News Brand Studio, the brand marketing arm of CBS News, were involved in the creation of this content. Throughout history, new forms of energy have shifted the gears of human progress, profoundly changing how people live and work. Ancient societies started with muscle, moved to animals, then harnessed water, steam and eventually, electricity. Today, AI is emerging as our next energy source, like fire, electricity or steam in past revolutions, to power productivity in the knowledge economy. Horses, steam engines and power plants all still matter of course, but the world is increasingly running on code and software. What some are calling Economy 4.0 is the AI-powered transformation of the knowledge economy, with software reordering how value is created, delivered and scaled. AI is the new power for business, but software is the power grid. It's the axle around which our knowledge economy is starting to turn. Software now shapes how we communicate, manage money, access healthcare, shop, learn, and nearly every other aspect of our modern lives. "Software is one of the key elements of making WellSky successful," said Joel Dolisy, Chief Technology Officer. "There is a lot of people interaction that is all enabled through software. [It] is a big, big important part of how we make that happen." And yet, much of the code that shapes the economy is old enough to collect social security. Nearly $3 trillion worth of global business operations still run on COBOL daily. Despite the increasing criticality of code, the world is still reliant on software that may have been developed by our grandparents. The impact: a chilling effect on disruptive innovation. Vimala Suresh, Vice President, CareSource, said, "If that file doesn't get processed in time, the member can't get the surgery they're eligible for. This could be life-threatening. That's how important software is in healthcare." Software development methods, processes, tools and mindsets from the past are crumbling under the pressure of AI-powered systems and potential. This presents leaders with a once-in-a-career opportunity. That's where Ascendion comes in. The company says it's unleashing intelligent agents that enhance humans as it builds software faster and better. Now, engineers using AAVA, a platform of agents aligned to software development and management, are using AI to deliver better code at half the cost and twice the speed. To unlock the potential value, leaders, Ascendion says IT teams, operations, line-of-business, procurement, finance and the C-suite must lean into new ways of designing, building and managing software. This isn't theory or buzzwords. It's happening now using what Ascendion calls Engineering to the Power of AI (EngineeringAI). This means companies can do more with less, allowing more time for talented humans to create, innovate, imagine and lead. David Park, President, Commercial Bank, Axos Bank, said, "With the speed of development and the reduction of cost through AI, we are able to personalize the banking experience. Hyper-personalization is what's most exciting about financial services and the software evolution we are going through." Marc Andreessen was right when he said, "Software is eating the world." Now, thanks to AI, software engineering is the first course. The data is adding up. Gartner predicts AI software will reach $297 billion by 2027. Researchers are uncovering a double-digit decrease in jobs related to writing and coding. A growing number of early adopter CIOs are already reporting that AI can yield efficiency gains of 20% to 40%. Innovation, growth, savings and transparency aren't new, but this is: AI is moving from the computer science lab to every business. The new power revolution from AI is already starting. The first major disruption where real value is being delivered today comes from how software is being built. AI is changing not just what is built, but how it's built. Every new power revolution—fire, water wheels, steam, electricity and now AI—comes with risks, disruptions and unintended consequences. AI's impact on software engineering will be profound, and it needs to be managed so we advance safely and civilly. In the coming years and decades, AI agents will disrupt countless business processes, consumer experiences and jobs. But today, software engineering is being re-coded by the bot. Business leaders, policy makers and employees can embrace the potential for AI and tame new machines so that value generation tips the scales away from risk, fear and failed projects.
Yahoo
an hour ago
- Yahoo
Ready to Start Exercising After Baby? Try This Postpartum Workout Plan
During pregnancy, eating a healthy diet and staying active is a top priority. But once baby is here, many women put their own health on the back burner. Even for those who want to prioritize movement, postpartum exercise can feel a bit overwhelming. Where should you start? How can you rebuild strength safely? The most important thing is to start slow and go at your own pace. Walking is a great low-impact activity you can do with your baby to ease into movement, boost your metabolism and improve cardiovascular health. An added bonus: On days that you are unable to squeeze in a full workout, going for a walk while pushing your baby is a great way to sneak in some exercise and allow for extra bonding time, too. When you're cleared for exercise by a doctor and ready to commit to a consistent workout plan, here are the four types of movements to add into your weekly workout routine. Benefits of postpartum exercise Exercising after having a baby is important for reasons far beyond fitting into your pre-pregnancy clothes. It's about reclaiming your energy, strength and overall well-being. Think of it as refueling your tank so you can be the best version of yourself for your baby. In terms of your physical health, getting back into a workout routine can help with strengthening your core to support your back, which is much-needed when you are constantly lifting and carrying your baby. Postpartum exercise can also boost your mood and reduce the risk of postpartum depression while improving sleep quality so you can better handle those sleepless nights. Postpartum workout plan A well-rounded workout plan includes core work, cardio, strength training and stretching. Here are some great exercises in each category that you can start incorporating into your routine. Always consult your doctor and wait until you are cleared for exercise before jumping into a fitness routine. Core These exercises can be done everyday to start rebuilding core strength and toning the midsection. Pelvic tilts Lying on your back with your feet on the floor, open your feet as wide as your hips. Reach your arms toward your heels. Take a deep breath in through your nose and fill your stomach up with air. Then, exhale through your mouth and pull your naval in toward your spine. To exaggerate this, as you breathe in slowly arch your low back up off of the ground, keeping your tailbone planted. Then, as you exhale press your low back into the ground as you tilt your pelvis towards your ribs. Repeat these pelvic tilts forward and back 10 times. Glute Bridge Lie flat on your back with your knees bent and feet flat on the ground. Your knees should be hip-distance apart. Perform a pelvic tilt and then raise your hips off the ground by squeezing your glutes to create a straight line from your neck to your knees, being careful not to hyperextend your hips. Pause for a breath before lowering back down, starting with the upper back and lowering each vertebrae until your butt is resting on the ground. Repeat 10 times. Dead Bug Begin by lying down flat on your back. Ensure your shoulders and low back remain flat on the floor. Next, lift your hands above your shoulders. Then, lift your legs and bend your knees, with your knees over your hips at a 90-degree angle. Exhale, engaging the core, as you slowly lower your opposite arm and leg (left leg and right arm), hovering them above the floor. Return to the starting position. Repeat on the opposite side (right leg and left arm). Repeat for 5 repetitions on each side. Half roll down Seated on the ground, bend your knees. (For a more advanced exercise, straighten your legs.) Then, reach your arms out in front of you with the shoulders relaxed. Pull the naval in toward the spine and slowly roll down half way to the floor. Hold for a breath and then slowly roll back up to the starting position. Repeat this ab exercise 10 times. Strength These exercises can be done every other day, but I recommend aiming for 3 times per week to start building full-body strength, boost metabolism and toning your muscles. I recommend performing three sets of 10 repetitions. Squats Stand with your feet shoulder-width apart and your toes pointed forward. Pull your navel in toward your spine to engage your core. Bend at the hips and knees while keeping your heels and toes on the floor. Slowly sit back into a squat position with your chest up, your shoulders back and abs in. Make sure that your knees are not crossing over your toes, and that you are as close to a 90-degree angle as possible. Straighten your legs by pressing into your heels to stand back up. Squeeze your glutes at the top, tilting your pelvis forward. Repeat. Backward lunge Stand with your feet as wide as your hips. Step the right foot back into a lunge position. Engage your abs and lower your right leg down until your knee almost touches the floor. Both knees should be at about a 90-degree angle. Push down through the front heel to come back up to standing. Repeat 10 times before switching sides. Bicep curls Stand tall with your feet hip-width apart. Engage your abs by tucking your pelvis and softly bend your knees. Hold one dumbbell in each hand (I recommend starting with 5-pound weights), and relax your arms so that they hang at the sides of your body. Make sure your palms are facing forward. Keep your shoulders back and down. Bend at your elbow and curl the dumbbells up toward your shoulders. Make sure to keep your elbows hugging the sides of your body. Lower both weights back down slowly. Exhale while lifting the weights, and inhale when you lower them down. Repeat 10 times. Tricep kickbacks Stand with your feet as wide as your hips, knees slightly bent. Hold on to one weight in each hand, and let your arms hang down toward your knees. Hinge at the hips so that your chest is tilted down toward the ground. Hug your elbows into your sides and pull your elbows up into a row. From here, keep the elbows stationary and extend the forearms back into the tricep kickback. Come back to the row position and then repeat the kickbacks. Repeat. Cardio No matter where you are in your fitness journey, a daily walking is something I encourage. Even on days when the weather - or your baby - doesn't cooperate you can get your steps in from the comfort of your living room by walking in place, doing laps around the house or doing one of my Indoor Walking Workout Routines. Other low-impact, postpartum-friendly cardio options include swimming, biking and rowing. Walking in place Walking in place is just as simple as it sounds — walk in place for 60 seconds bursts. Swing your arms, smile and breathe in and out through your nose. For an added challenge, walk with high knees, pulling one knee up to hip height (forming a 90-degree angle) before lowering it to the ground and raising the opposite knee. Stretch Make stretching a part of your daily routine. You can use these exercises as a warmup or cool down for a workout or simply sprinkle them throughout the day to help ease tension and combat aches and pains. Wide-leg forward fold Stand in a wide straddle with your feet parallel. Bring your hands to your hips or let them hang by your sides. Inhale and look up to the sky. Exhale and hinge at the hips, folding forward. Keep the hips in line with the heels and reach your fingertips toward the ground. If your flexibility allows, walk your fingertips forward while you pull your hips back. Ground through the heels. Hold for a few breaths and then slowly roll back up to standing. Repeat. Cat Cow Start on all fours with your hands below your shoulders and knees below your hips. Breathe in while lifting your head up toward the ceiling. Arch your back as you breathe in, moving your stomach toward the ground. Then, exhale while pulling your chin to your chest and pulling your abs in so that you arch your back toward the ceiling. Continue alternating back and forth for about one minute. Crescent Lunge From a standing position, step your left foot back behind you. Bend your right knee to a 90-degree angle, keeping your knee over your toes. Reach your arms toward the sky and straighten your back. Hold for 5 breaths, relaxing the shoulders, before switching sides. Hamstring stretch Start standing tall. Step your right foot 1-2 feet forward and bend your left knee as you lean forward over the straight right leg. Flex the right foot. Feel a stretch down the back of your leg and into your calf. Hold for 2-4 breaths and then switch sides. This article was originally published on
Forbes
2 hours ago
- Forbes
Analysis Of Whether Generic Generative AI Falls Within The Purview Of Providing Therapy And Psychotherapeutic Advice
In today's column, I examine a seemingly straightforward question that asks whether contemporary generic generative AI and large language models (LLMs) are said to be providing therapy and psychotherapeutic advice. The deal is this. When you use ChatGPT, Claude, Llama, Gemini, Grok, and other such popular generative AI systems, you can readily engage the AI in conversations about mental health. This can be of a general nature. It can also be a very personal dialogue. Many people are using AI as their de facto therapist and doing so without nary a thought of reaching out to a human therapist or mental health professional. Does the use of those LLMs in this manner signify that the AI is proffering services constituting therapy and psychotherapy? You might declare that yes, of course, that is precisely what the AI is doing. It is blatantly obvious. But AI makers who make and maintain the AI are undoubtedly reluctant to agree with that plain-stated assessment or ad hoc opinion. You see, new laws are starting to be enacted that bear down on generic AI that provides unfettered services within the scope of therapy and psychotherapy. AI makers are likely to desperately contend that their generic AI falls outside that regulatory scope. The question arises whether they will be successful in making that kind of tortuous argument. Some would say they don't have a ghost of a chance. Others believe they can dance their way around the legally troubling matter and come out scot-free. Let's talk about it. This analysis of AI breakthroughs is part of my ongoing Forbes column coverage on the latest in AI, including identifying and explaining various impactful AI complexities (see the link here). AI And Mental Health Therapy As a quick background, I've been extensively covering and analyzing a myriad of facets regarding the advent of modern-era AI that produces mental health advice and performs AI-driven therapy. This rising use of AI has principally been spurred by the evolving advances and widespread adoption of generative AI. For a quick summary of some of my posted columns on this evolving topic, see the link here, which briefly recaps about forty of the over one hundred column postings that I've made on the subject. There is little doubt that this is a rapidly developing field and that there are tremendous upsides to be had, but at the same time, regrettably, hidden risks and outright gotchas come into these endeavors too. I frequently speak up about these pressing matters, including in an appearance last year on an episode of CBS's 60 Minutes, see the link here. State Law Ups The Ante I recently analyzed a newly enacted law on AI for mental health that had been signed and enacted in Illinois on August 1, 2025, see my coverage at the link here. This new law is quite a doozy. The reason that it is a doozy is that it lays out violations and penalties for AI that provides unfettered therapy and psychotherapy services. The implication is that any generic generative AI, such as the popular ones I noted earlier, is now subject to potential legal troubles. Admittedly, the legal troubles right now would seemingly be confined to aspects of or within Illinois, since this is a state law and not a broader federal law. Nonetheless, in theory, the use of generic generative AI by users in Illinois that, by happenstance, provides therapy or psychotherapeutic advice is presumably within the scope of getting dinged by the new law. You can bet your bottom dollar that similar new laws are going to be popping up in many other states. The clock is ticking. And the odds are that this type of legislation will also spur action in the U.S. Congress and potentially lead to federal laws of a like nature. It all could have a tremendous impact on AI makers, along with major impacts on how generative AI is devised and made available to the public. All in all, few realize the significance of this otherwise innocuous and under-the-radar concern. My view is that this is the first tiny snowball that is starting to roll down a snowy hill and soon will be a gigantic avalanche that everybody will be talking about. Time will tell. Background On AI For Mental Health I'd like to set the stage before we get into the particulars of this heady topic. You might be vaguely aware that the top-ranked public use of generative AI and LLMs is to consult with the AI on mental health considerations, see my coverage at the link here. This makes abundant sense. You can access most of the major generative AI systems for nearly free or at a super low cost, doing so anywhere and at any time. Thus, if you have any mental health qualms that you want to chat about, all you need to do is log in to AI and proceed forthwith on a 24/7 basis. Compared to using a human therapist, the AI usage is a breeze and readily undertaken. AI makers already find themselves in a bit of a pickle on this usage of their AI. The deal is this. By allowing their AI to be used for mental health purposes, they are opening the door to legal liability if their AI gets caught dispensing inappropriate guidance and someone suffers harm accordingly. So far, AI makers have been relatively lucky and have not yet gotten severely stung by their AI serving in a therapist role. You might wonder why the AI makers don't just shut off the capability of their AI to produce mental health insights. That would solve the problem of the business exposures involved. Well, as noted above, this is the top attractor for people to use generative AI. It would be usurping the cash cow, or like capping an oil well that is gushing out liquid gold. One aspect that the AI makers have already undertaken is to emphasize in their online licensing agreements that users aren't supposed to use the AI for mental health advice, see my coverage at the link here. The aim is that by telling users not to use the AI in this manner, perhaps the AI maker can shield itself from adverse exposure. The thing is, despite the warnings, the AI makers often do whatever they can to essentially encourage or support the use of their AI for this claimed-to-be don't use capacity. Some would insist this is a wink-wink of trying to play both sides of the gambit at the same time, see my discussion at the link here. The Services Question My commentary on these sobering matters is merely a layman's viewpoint. Make sure to consult with your attorney to garner any legal ramifications pertaining to your situation and any potential legal entanglements regarding AI and mental health. Let's take a look at the Illinois law that was recently passed. According to the Wellness and Oversight for Psychological Resources Act, known as HB1806, these two elements are a core consideration (excerpts): Regarding the use of unregulated AI in this realm, a crucial statement about AI usage for mental health purposes is stated this way in the Act (excerpt): There are varying ways to interpret this wording. One interpretation is that if an AI maker has a generic generative AI that also happens to provide mental health advice, and if this is taking place without the supervision of a licensed professional, and this occurs in Illinois, the AI maker is seemingly in violation of this law. The AI maker might not even be advertising that their AI can be used that way, but all it takes is for the AI to act in such a manner (since it provides or offers as such). Generic AI Versus Purpose-Built AI Closely observe that the new law stipulates that the scope involves 'therapy or psychotherapy services'. This brings us back to my opening question: Before we unpack the thorny issue, I'd like to clarify something about the topic of AI for mental health. You might have noticed that I referred to generic generative AI. What does the word 'generic' mean in this context? Let me explain. Well, first, there are customized generative AI systems and AI-based apps that are devised specifically to carry out mental health activities. Those are specially built for that purpose. It is the obvious and clear-cut intent of the AI developer that they want their AI to be used that way, including that they are likely to advertise and promote the AI for said usage. See my coverage on such purpose-built AI for mental health at the link here and the link here. In contrast, there is generic generative AI that just so happens to have a capability that encompasses providing mental health advisement. Generic generative AI is intended to answer all kinds of questions and delve into just about any topic under the sun. The AI wasn't especially tuned or customized to support mental health guidance. It just happens to be able to do so. I am focusing here on the generic generative AI aspects. The custom-built AI entails somewhat similar concerns but has its own distinct considerations. I'll be going into those facets in an upcoming posting, so be on the watch. Definitions And Meaning Are Crucial An AI maker might claim that they aren't offering therapy or psychotherapy services and that their generic generative AI has nothing to do with therapy or psychotherapy services. It is merely AI that interacts with people on a wide variety of topics. Period, end of story. The likely retort is that if your AI is giving out mental health advice, it falls within the rubric of therapy and psychotherapy services (attorneys will have a field day on this). Thus, trying to dodge the law by being sneaky about wording isn't going to get you off the hook. If it walks like a duck and quacks like a duck, by gosh, it surely is a duck. One angle on this disparity or dispute would be to nail down what the meaning and scope of therapy and psychotherapy encompass. Before we look at what the Illinois law says, it is useful to consider definitions from a variety of informed sources. Definitions At Hand According to the online dictionary of the American Psychological Association (APA), therapy and psychotherapy are defined this way: The Mayo Clinic provides this online definition: The National Institute Of Health (NIH) provides this online definition: And, the popular website and publication Psychology Today has this online definition: Interpreting The Meanings Those somewhat informal definitions seem to suggest that the nature of therapy and psychotherapy includes these notable elements: (1) aiding mental health problems, (2) use 'talk' or interactive chatting as a mode of communication, and (3) undertaken by a mental health professional. Let's see what the Illinois law says about therapy and psychotherapy (excerpts per the Act): It is interesting and notable that some carve-outs were made. The scope appears to exclude peer support, along with excluding religious counseling. Contemplating The Matter It might be worthwhile to noodle on how an AI maker might seek to avoid repercussions from their generic generative AI getting caught up in this messy milieu. First, if therapy and psychotherapy were defined as requiring that a mental health professional be involved, this provides an angle of escape. Why so? Oddly enough, an AI maker could simply point out that their AI doesn't employ or otherwise make use of a mental health professional. Therefore, the AI cannot be providing these said services since it fails to incorporate a supposed requirement. Notably, the Illinois law seems not to fall into that trap, since it seems to simply indicate that there are services and does not name that a mental health professional is part and parcel of the definition. Some of the other definitions that I listed would potentially be in a murkier condition due to explicitly mentioning a required role of having a trained professional or other similar verbiage. Second, an AI maker might try to claim that their generic generative AI is more akin to peer support. The beauty there is that since peer support is a carve-out, perhaps their AI is no longer within scope. It would be a tough row to hoe. Peer support stipulates that individuals are involved. At this juncture, we do not genuinely recognize AI as having legal personhood, see my discussion at the link here, and therefore, trying to assert that AI is an 'individual' would be an extraordinary stretch. Third, an AI maker might go the route of claiming that their generic generative AI is a form of religious counseling. The advantage would be that the matter of religious consulting is a carve-out. In that case, if AI were said to be doing religious counseling when providing mental health advice, the AI maker would apparently be free of the constraint. This appears to be a failing strategy for several reasons, including that the AI is presumably not a clergy member, pastoral counselor, or other religious leader (maybe a desperate attempt could be made to anoint the AI in that fashion, but this would seem readily overturned). Caught In A Web Other potential dodges or efforts to skirt the coming set of laws will indubitably be a keen topic for legal beagles and legal scholars. If an AI maker doesn't find a viable workaround, they are going to be subject to various fines and penalties. Those could add up. For example, Illinois has a population of approximately twelve million people. Of those, suppose that half are using generic generative AI (that's a wild guess), and that half of those use the AI for mental health aspects from time to time (another wild guess). That would be three million people, and each time they use the AI for that purpose might be construed as a violation. If each person does so once per week, that's twelve million violations in a month. The Illinois law says that each violation is up to a maximum fine of $10,000. We'll imagine that instead of the maximum, an AI maker gets fined a modest $1,000 per violation. In one month, based on this spitball conjecture, that could be $12 billion in fines. Even the richest tech firms are going to pay attention to that kind of fine. Plus, once other states go the same route, you can multiply this by bigger numbers for each of the additional states and how they opt to penalize AI that goes over the line. Crucial Juncture At Hand An ongoing and vociferously heated debate concerns whether the use of generic generative AI for mental health advisement on a population-level basis is going to be a positive outcome or a negative outcome for society. If that kind of AI can do a proper job on this monumental task, then the world will be a lot better off. You see, many people cannot otherwise afford or gain access to human therapists, but access to generic generative AI is generally plentiful in comparison. It could be that such AI will greatly benefit the mental status of humankind. A dour counterargument is that such AI might be the worst destroyer of mental health in the history of humanity. See my analysis of the potential widespread impacts at the link here. So far, AI makers have generally had free rein with their generic generative AI. It seems that the proverbial rooster has finally headed home to roost. Gradually, new laws are going to be enacted that seek to prohibit generic generative AI from dispensing mental health advice that is absent of a human therapist performing counseling. Get yourself primed and ready for quite a royal battle that might determine the future mental status of us all.
