“That is a pretty gnarly chest X-ray and I would not want to have that disease.”
Minnesota medical resident and social influencer, Dr. Rose Marie Leslie, describing an X-ray of an e-cigarette user’s lungs in her viral TikTok video. This may not be the best approach for patient consultations, but it proved to be effective for TikTok.
Imaging Wire Sponsors
- Carestream – Focused on delivering innovation that is life changing – for patients, customers, employees, communities and other stakeholders
- Focused Ultrasound Foundation – Accelerating the development and adoption of focused ultrasound
- Medmo – Helping underinsured Americans save on medical scans by connecting them to imaging providers with unfilled schedule time
- Nuance – AI and cloud-powered technology solutions to help radiologists stay focused, move quickly, and work smarter
- Pocus Systems – A new Point of Care Ultrasound startup, combining a team of POCUS veterans with next-generation genuine AI technology to disrupt the industry
- Qure.ai – Making healthcare more accessible by applying deep learning to radiology imaging
The Imaging Wire
Screening at 30
Research published in JACR found that women with a family history of breast cancer, a personal history of breast cancer, and/or dense breasts would likely benefit from mammography screening starting at age 30, not 40.
If this Sounds Familiar – It’s because the same team shared results from an earlier version of this study at RSNA 2018, leading to some debates about age vs. risk screening qualifications and criticisms of the current screening standards.
Updated Study – Using data from 5.98m screening mammograms (2.64m women), the team calculated four screening performance metrics (recall rate, cancer detection rate, PPV for biopsy recommended, and PPV for biopsy performed) for each risk group and 5yr age group.
Similar Results – The study found that 30-39yr women with at least one of the above risk factors had similar diagnostic metrics as 40-49yr women with average risk factors, who are already recommended for screening in the US (suggesting that they are equally qualified for screening). Additionally, women with these risk factors in the 30-34yr age group had similar diagnostic metrics as at-risk women in the 35-39yr group (suggesting that screening should start at 30).
The Case for Past Imaging History
An editorial from an Italian team declared that “Past Imaging History” should be a standard part of patients’ medical history. Here’s some support for that:
Stuck in History – The article called for an evolution in medical histories and PACS usage, arguing that “the way a medical history is taken hasn’t changed a lot” for decades, and even with the growth of multi-modality imaging, PACS remains “underrated and not fully exploited.”
Adopting Past Imaging History – The team believes making PACS with Past Imaging History available to all clinicians has “enormous potential” to reveal important information that might not otherwise be communicated to clinicians (e.g. reveal presence of prosthetic materials or the date of previous procedures) and help avoid unnecessary exams.
Some may not be as bullish about the short-term technological/procedural feasibility of making Past Imaging History a standard, and some might not love making this information available to all clinicians, but it’s harder to argue against the fact that there are millions of historical images/reports going largely unused and unreferenced.
There’s been a lot of imaging-related coverage of vaping’s effects on the lungs lately, but a viral TikTok video detailing vapers’ chest X-rays might have the greatest public awareness impact we’ve seen to-date. Here’s how it went down:
Going Viral – A video from Dr. Rose Marie Leslie ( 2nd-year, Minnesota’s North Memorial Health Hospital) comparing healthy and vaper chest X-rays generated 570k likes and 3.5 million views (probably more by now), while gaining coverage across a wide range of mainstream media.
Simple & Serious – Dr. Leslie pointed out how the vaper’s X-ray showed signs of inflammation, fluid, infection, and/or bleeding, describing it as “pretty gnarly.” There’s a medical way to describe these symptoms, but Dr. Leslie’s conversational and social-savvy style allowed this video to have far more influence on TikTok’s young audience than the dozens of articles we’ve seen in mainstream and medical pubs over the last few months. It’s probably the first time a lot of these >3.5 million viewers got a glimpse of how an X-ray is interpreted, too.
- An article from the University of Pennsylvania and University of Washington found that adding a default order to EHR systems (aka a “nudge”) can improve clinician decision making for imaging-based cancer screening and help them avoid “predictably irrational behaviors” (e.g. prioritizing short-term factors or overweighting low probability events). The team outlined the types of nudges that can be used (e.g. encourage discussion, follow guidelines, automatically schedule appointments, encourage patient screening adherence) and urged healthcare organizations and radiologists to work to implement EHR nudges. This UPenn team is becoming a nudge research leader, as it previously used the method to cut unnecessary imaging for palliative radiation therapy in half.
- Guerbet and IBM Watson Health expanded their year-old strategic AI partnership, announcing plans to co-develop and commercialize a new AI tool that helps radiologists/oncologists detect, segment, characterize, and monitor prostate cancer lesions. The new prostate cancer tool is designed for PACS compatibility and will leverage technology developed in Guerbet and IBM’s previously-launched liver cancer diagnostic solution.
- Over a year after Australia/New Zealand’s RANZCR took a stand against radiographer image interpretations, an Australian team published a review of all of the country’s related studies (n=19) to provide evidence-based support for this practice and detail how it can gain greater adoption. The review found a wide range of radiographer image interpretation performance (accuracy 57% – 98%, sensitivity 45% – 98%, and specificity 68% – 98%) and suggested that radiographers can indeed perform preliminary image evaluations (PIEs) if they were provided the right training,had better defined image evaluation roles, and greater clinical support.
- Medical imaging AI startup Kheiron Medical Technologies announced a $22 million series A round that it will use to fund large-scale clinical trials on its Mia (mammography intelligent assessment) breast cancer screening software. Mia launched in Europe earlier this year and is awaiting FDA clearance.
- Walmart sent another message that it’s serious about its healthcare expansion, adding healthcare to its Live Better U employee education program (this costs employees just $1 a day). None of Walmart’s healthcare education fields directly overlap with our kind of imaging, but the retail giant’s healthcare push is worth everyone’s attention and these homegrown healthcare professionals appear to be a core part of Walmart’s health clinic growth strategy.
- Swedish biopharmaceutical company Affibody announced a global partnership with GE Healthcare to develop and commercialize Affibody-based PET imaging tracers, initially focused on HER2 (for metastatic breast cancer) and PD-L1 (immuno-oncology treatment patient selection). The announcement comes just a week after GE and AdAlta announced a similar PET tracer development alliance, which appears to be more than a coincidence given that GE stated in this latest announcement that it is relying on partnerships like this “to build a portfolio of molecular imaging agents for oncology.”
- The ethical AI debate made it all the way to Fast Company, which recently published “5 Simple Rules to Make AI a Force for Good.” Here they are as fast and simple as possible: 1) Create an FDA for algorithms; 2) Open up AI’s blackbox; 3) Value human wisdom over #AI wizardry; 4) Make privacy the default; 5) Compete by promoting, not infringing, civil rights.
- French healthcare AI company Incepto announced a €5.6 million series A round, calling the funding “a first step in accelerating the large-scale rollout of [its] solutions throughout France” and towards it goal of becoming one of Europe’s medical imaging AI leaders. Although imaging AI funding announcements are pretty common these days, Incepto gets credit for having a unique strategy. Incepto intends to build a “Netflix of AI” platform that connects AI companies with hospitals for co-development and positions Incepto as both an AI developer and distributor (it already distributes solutions from Screenpoint and Qure.ai).
- Clarius Mobile Health announced a partnership with Savvik Buying Group, making it the emergency medicine non-profit’s only approved ultrasound vendor and introducing Clarius to Savvik’s 11,600 members (e.g. ambulance services, police, fire).
- MITA, the ACR, and a list of medtech heavyweights teamed up for one last big message, asking congress to repeal the 2.3% medical device tax before its latest 2-year suspension expires on January 1, 2020 (previously suspended 2016-17, 2018-19). The letter attributed thousands of lost jobs to the tax’s 2013-2015 active run and warned that it could lead to another 21,000 job cuts if it’s reinstated.
- The Mount Sinai Health System introduced its new Biomedical Engineering and Imaging Institute (BMEII), representing a merger and expansion of its imaging, nanomedicine, and biomedical engineering programs. Launching in early 2020, the Mount Sinai BMEII will staff at least nine principal investigators and student teams focused on developing novel technologies in 1) Artificial Intelligence in Advanced Imaging; 2) Next Generation Medical Technologies; and 3) Virtual, Augmented, and Extended Reality (VR/AR/XR).
- South Korean flat-panel DR developer, Rayence, launched a new Super IGZO-based high-speed video X-ray detector. The new detector overcomes the short lifespan that IGZO-based panels are known for with a new design that reportedly gives it 20 times larger radiation hardness than other IGZO panels, while also reducing radiation dosage by 30%.
The Resource Wire
– This is sponsored content.
- Qure.ai CCO, Chiranjiv Singh, shares about the importance of building trust in healthcare AI in this editorial.
- This Carestream case study compares images of foot trauma captured using the OnSight 3D Extremity System to images captured on 2D X-rays.
- Nuance’s Karen Holzberger will take the stage at this week’s Biennial of The Americas Festival in Denver to discuss empathy in artificial intelligence.
- Focused ultrasound researchers at the University of Michigan were awarded a $2.4M grant from the National Institute of Biomedical Imaging and Bioengineering (NIBIB) to study using histotripsy for brain applications. Histotripsy is a form of focused ultrasound that mechanically destroys tissue without heat.
- Did you know that imaging patients are most likely to no-show for their procedures on Mondays and Saturdays? By partnering with Medmo, imaging centers can keep their schedules full, despite the inevitable Monday no-shows.
- POCUS Systems’ forthcoming ultrasounds will combine ease of use, durability, and reliability, allowing clinicians to focus on their patients.