Latest news with #TFM
Medscape
4 days ago
- Health
- Medscape
AI Algorithm Predicts Transfusion Need in Acute GI Bleeds
SAN DIEGO — A novel generative artificial intelligence (AI) framework known as trajectory flow matching (TFM) can predict the need for red blood cell transfusion and mortality risk in intensive care unit (ICU) patients with acute gastrointestinal (GI) bleeding, researchers reported at Digestive Disease Week (DDW) 2025. Acute GI bleeding is the most common cause of digestive disease–related hospitalization, with an estimated 500,000 hospital admissions annually. It's known that predicting the need for red blood cell transfusion in the first 24 hours may improve resuscitation and decrease both morbidity and mortality. However, an existing clinical score known as the Rockall Score does not perform well for predicting mortality, Xi (Nicole) Zhang, an MD-PhD student at McGill University, Montreal, Quebec, Canada, told attendees at DDW. With an area under the curve of 0.65-0.75, better prediction is needed, Zhang said, whose coresearchers included Dennis Shung, MD, MHS, PhD, assistant professor of medicine and director of Applied Artificial Intelligence at Yale University School of Medicine, New Haven, Connecticut. 'We'd like to predict multiple outcomes in addition to mortality,' said Zhang, who is also a student at the Mila-Quebec Artificial Intelligence Institute. As a result, the researchers turned to the TFM approach, applying it to ICU patients with acute GI bleeding to predict both the need for transfusion and in-hospital mortality risk. The all-cause mortality rate is up to 11%, according to a 2020 study by James Y. W. Lau, MD, and colleagues. The rebleeding rate of nonvariceal upper GI bleeds is up to 10.4%. Zhang said the rebleeding rate for variceal upper gastrointestinal bleeding is up to 65%. The AI method the researchers used outperformed a standard deep learning model at predicting the need for transfusion and estimating mortality risk. Defining the AI Framework 'Probabilistic flow matching is a class of generative artificial intelligence that learns how a simple distribution becomes a more complex distribution with ordinary differential equations,' Zhang told Medscape Medical News. 'For example, if you had a few lines and shapes you could learn how it could become a detailed portrait of a face. In our case, we start with a few blood pressure and heart rate measurements and learn the pattern of blood pressures and heart rates over time, particularly if they reflect clinical deterioration with hemodynamic instability.' Another way to think about the underlying algorithm, Zhang said, is to think about a river with boats where the river flow determines where the boats end up. 'We are trying to direct the boat to the correct dock by adjusting the flow of water in the canal. In this case we are mapping the distribution with the first few data points to the distribution with the entire patient trajectory.' The information gained, she said, could be helpful in timing endoscopic evaluation or allocating red blood cell products for emergent transfusion. Study Details The researchers evaluated a cohort of 2602 patients admitted to the ICU, identified from the publicly available MIMIC-III database. They divided the patients into a training set of 2342 patients and an internal validation set of 260 patients. Input variables were severe liver disease comorbidity, administration of vasopressor medications, mean arterial blood pressure, and heart rate over the first 24 hours. Excluded was hemoglobin, since the point was to test the trajectory of hemodynamic parameters independent of hemoglobin thresholds used to guide red blood cell transfusion. The outcome measures were administration of packed red blood cell transfusion within 24 hours and all-cause hospital mortality. The TFM was more accurate than a standard deep learning model in predicting red blood cell transfusion, with an accuracy of 93.6% vs 43.2%; P ≤ .001. It was also more accurate at predicting all-cause in-hospital mortality, with an accuracy of 89.5% vs 42.5%, P = .01. The researchers concluded that the TFM approach was able to predict the hemodynamic trajectories of patients with acute GI bleeding defined as deviation and outperformed the baseline from the measured mean arterial pressure and heart rate. Expert Perspective 'This is an exciting proof-of-concept study that shows generative AI methods may be applied to complex datasets in order to improve on our current predictive models and improve patient care,' said Jeremy Glissen Brown, MD, MSc, an assistant professor of medicine and a practicing gastroenterologist at Duke University who has published research on the use of AI in clinical practice. He reviewed the study for Medscape Medical News but was not involved in the research. 'Future work will likely look into the implementation of a version of this model on real-time data.' He added: 'We are at an exciting inflection point in predictive models within GI and clinical medicine. Predictive models based on deep learning and generative AI hold the promise of improving how we predict and treat disease states, but the excitement being generated with studies such as this needs to be balanced with the trade-offs inherent to the current paradigm of deep learning and generative models compared to more traditional regression-based models. These include many of the same 'black box' explainability questions that have risen in the age of convolutional neural networks as well as some method-specific questions due to the continuous and implicit nature of TFM.' Elaborating on that, Glissen Brown said: 'TFM, like many deep learning techniques, raises concerns about explainability that we've long seen with convolutional neural networks — the 'black box' problem, where it's difficult to interpret exactly how and why the model arrives at a particular decision. But TFM also introduces unique challenges due to its continuous and implicit formulation. Since it often learns flows without explicitly defining intermediate representations or steps, it can be harder to trace the logic or pathways it uses to connect inputs to outputs. This makes standard interpretability tools less effective and calls for new techniques tailored to these continuous architectures.' 'This approach could have a real clinical impact,' said Robert Hirten, MD, associate professor of medicine and artificial intelligence, Icahn School of Medicine at Mount Sinai, New York City, who also reviewed the study. 'Accurately predicting transfusion needs and mortality risk in real time could support earlier, more targeted interventions for high-risk patients. While these findings still need to be validated in prospective studies, it could enhance ICU decision-making and resource allocation.' 'For the practicing gastroenterologist, we envision this system could help them figure out when to perform endoscopy in a patient admitted with acute gastrointestinal bleeding in the ICU at very high risk of exsanguination,' Zhang told Medscape Medical News. The approach, the researchers said, will be useful in identifying unique patient characteristics, make possible the identification of high-risk patients and lead to more personalized medicine. Hirten, Zhang, and Shung had no disclosures. Glissen Brown reported consulting relationships with Medtronic, OdinVision, Doximity, and Olympus.
Economic Times
27-05-2025
- Science
- Economic Times
What's slithering beneath the Great Lakes has 150 teeth, a serrated tongue, and a ruthless hunger that could collapse the entire ecosystem
Canadian and US agencies are collaborating to combat the invasive sea lamprey in the Great Lakes, aiming to protect the $5.1-billion fishery. Control measures, including lampricides and barriers, are being implemented after a surge in lamprey populations due to pandemic-related disruptions and staffing cuts. Sustained efforts are crucial to prevent further damage to native fish and the regional economy. Tired of too many ads? Remove Ads What are sea lampreys? Tired of too many ads? Remove Ads The 'Trump' factor The invasive sea lamprey , often dubbed 'vampire fish" for its bloodsucking predation, has prompted Canadian and US agencies to launch a joint campaign to protect the Great Lakes ' $5.1-billion freshwater crews initiated treatments in late April, focusing on Lake Erie and Lake Ontario , with operations expected to continue through methods employed include lampricides (chemical treatments), traps, and barriers aimed at reducing the sea lamprey population, which parasitically feeds on lampreys, native to the Atlantic Ocean, were inadvertently introduced into the Great Lakes via shipping canals in 1921. Lacking natural predators in this new environment, they have wreaked havoc on native fish populations. Using their mouths, equipped with 150 teeth and a serrated tongue, they latch onto prey, often killing up to 40 pounds of fish during their parasitic COVID-19 pandemic exacerbated the issue. Travel restrictions in 2020 and 2021 forced crews to pause treatments, allowing millions of lamprey larvae to survive.A new study confirms that lamprey numbers rose sharply during this period. In 2024, adult sea lamprey counts were 8,619 above pre-COVID-19 averages, with Lake Superior and Lake Ontario experiencing the most significant Baker, chair of the Great Lakes Fishery Commission, emphasized the urgency:'Ongoing, consistent sea lamprey control is critically important for preventing damage to Great Lakes fish by invasive sea lampreys. If we take our foot off the gas, even for a short while, sea lamprey populations will increase rapidly and cause considerable damage to fish.'The Great Lakes Fishery Commission, established in 1955, has been instrumental in reducing sea lamprey populations by 90 percent in most areas of the Great Lakes. Their integrated control program combines several methods to attack sea lampreys on multiple primary method involves applying the lampricide TFM to target sea lamprey larvae in their nursery tributaries. In the concentrations used, TFM kills larvae before they develop lethal mouths and migrate to the lakes to feed on fish, while most other organisms are Trump administration's mass purge of federal workers, including 12 members of the Great Lakes sea lamprey control program, has raised concerns about the program's efficacy. These staffing cuts represent nearly 14 percent of the lamprey control program's overall workforce, potentially allowing over 1 million lamprey to survive, consuming nearly 5 million pounds of fish and resulting in $105 million in lost economic these setbacks, control efforts have resumed. Scientists and officials warn that the damage done during the pause will take years to Great Lakes Fishery Commission continues to stress the importance of sustained efforts to safeguard the prosperity of the region.
Time of India
27-05-2025
- Science
- Time of India
What's slithering beneath the Great Lakes has 150 teeth, a serrated tongue, and a ruthless hunger that could collapse the entire ecosystem
The invasive sea lamprey , often dubbed 'vampire fish" for its bloodsucking predation, has prompted Canadian and US agencies to launch a joint campaign to protect the Great Lakes ' $5.1-billion freshwater fishery. Control crews initiated treatments in late April, focusing on Lake Erie and Lake Ontario , with operations expected to continue through October. The methods employed include lampricides (chemical treatments), traps, and barriers aimed at reducing the sea lamprey population, which parasitically feeds on fish. What are sea lampreys? Sea lampreys, native to the Atlantic Ocean, were inadvertently introduced into the Great Lakes via shipping canals in 1921. Lacking natural predators in this new environment, they have wreaked havoc on native fish populations. Using their mouths, equipped with 150 teeth and a serrated tongue, they latch onto prey, often killing up to 40 pounds of fish during their parasitic stage. Live Events The COVID-19 pandemic exacerbated the issue. Travel restrictions in 2020 and 2021 forced crews to pause treatments, allowing millions of lamprey larvae to survive. A new study confirms that lamprey numbers rose sharply during this period. In 2024, adult sea lamprey counts were 8,619 above pre-COVID-19 averages, with Lake Superior and Lake Ontario experiencing the most significant increases. Ethan Baker, chair of the Great Lakes Fishery Commission, emphasized the urgency: 'Ongoing, consistent sea lamprey control is critically important for preventing damage to Great Lakes fish by invasive sea lampreys. If we take our foot off the gas, even for a short while, sea lamprey populations will increase rapidly and cause considerable damage to fish.' The Great Lakes Fishery Commission, established in 1955, has been instrumental in reducing sea lamprey populations by 90 percent in most areas of the Great Lakes. Their integrated control program combines several methods to attack sea lampreys on multiple fronts. The primary method involves applying the lampricide TFM to target sea lamprey larvae in their nursery tributaries. In the concentrations used, TFM kills larvae before they develop lethal mouths and migrate to the lakes to feed on fish, while most other organisms are unaffected. The 'Trump' factor The Trump administration's mass purge of federal workers, including 12 members of the Great Lakes sea lamprey control program, has raised concerns about the program's efficacy. These staffing cuts represent nearly 14 percent of the lamprey control program's overall workforce, potentially allowing over 1 million lamprey to survive, consuming nearly 5 million pounds of fish and resulting in $105 million in lost economic output. Despite these setbacks, control efforts have resumed. Scientists and officials warn that the damage done during the pause will take years to reverse. The Great Lakes Fishery Commission continues to stress the importance of sustained efforts to safeguard the prosperity of the region.
Yahoo
23-05-2025
- Yahoo
South Dakota agencies seek out ICE authority under 287(g)
SIOUX FALLS, S.D. (KELO) – Two more South Dakota law enforcement agencies are seeking out hands-on authority to identify and remove illegal immigrants from the state. The Division of Criminal Investigation (DCI) and Highway Patrol applied for a U.S. Immigration and Customs Enforcement (ICE) program called 287(g), which trains and certifies certain officers with specified immigration duties, with limitations and oversight depending on the agreement's terms. The two agencies both applied under the Task Force Model (TFM), where officers incorporate immigration authority in their day-to-day duties. Back in April, two South Dakota sheriff's offices joined 287(g) as warrant service officers (WSOs), who can identify criminal immigrants and immigration violators who are already in custody. As of May 22, these are the only two confirmed participating agencies from South Dakota under 287(g). If the DCI and Highway Patrol's applications are approved, then four total state law enforcement agencies will perform immigration authority tasks for ICE under 287(g). ICE activity increased in South Dakota, after agents arrested eight people in Madison at Manitou Equipment and Global Polymer Industries in May. Under the Immigration and Nationality Act (INA), section 287(g), ICE's Enforcement and Removal Operations (ERO) partner with law enforcement agencies under a memorandum of agreement (MOA) for specified immigration authority. Only law enforcement agencies can apply for ICE's 287(g) program. According to ICE, 287(g) operates under three main program models: Warrant Service Officer (WSO): ICE trains and certifies certain law enforcement officers only within the agency's jail to serve and execute administrative or civil immigration warrants on removable immigrants. Jail Enforcement Model (JEM): allows officers inside a jail to identify, interrogate, and determine immigration status of individuals, (who possess convicted criminal or pending criminal charges and been arrested by state or local law enforcement agencies), and aid in potential immigration proceedings. Task Force Model (TFM): a 'force multiplier,' intended for law enforcement officers on the field or on routine police duties to enforce limited immigration authority, with ICE oversight. While applying for the program, letters of intent and an MOA must be submitted to the U.S. Department of Homeland Security (DHS) for consideration. According to ICE, letters of intent must specify which model program will be used, the current relationship between the law enforcement agency and ICE, the law enforcement agency's location, and the number of deputies that will receive training. Once an agency signs an MOA for the 287(g) program, ICE will train law enforcement in immigration law, multicultural communication, and methods to avoid racial profiling. According to ICE's website, several states have passed or are considering state legislation mandating law enforcement agencies to apply for MOAs under 287(g). On Wednesday South Dakota Attorney General Marty Jackley announced that DCI agents will train as task force officers under 287(g). Two DCI officers- one from East river and one from West river- will train as task force officers under TFM, according to Jackley. As a task force officer under 287(g), the two agents will be granted immigration law enforcement as a part of their day-to-day duties. 'It's a force multiplier, it saves the taxpayers, arrives at the right solution- because it gets violent criminals and drug dealers- that are not legally- here out of our state,' Jackley said in a Friday interview with KELOLAND News. Jackley said he thinks the community will respond positively to the partnership because this federal authority will only apply to violent immigrant criminals and drug dealers. 'Nobody wants them in our community,' Jackley said. 'They're not here legally.' Jackley said TFM will speed up the process for state and federal authorities to directly move forward with deportations. 'Nobody wants to see a victim go through more process than the victim has to, or use of taxpayer dollars that's either duplicative or not effective,' Jackley said. Jackley said he hopes to train more officers in the future, with just under 70 agents working for the DCI. 'It's going to take a little time,' Jackley said. 'I want to get established where we have at least components in East river and West river have that leadership in place, and then as the need comes about, we will train additional agents.' According to Jackley's office, a Task Force Program will be headquartered in Pierre. Jackley announced the partnership between the DCI and ICE after touring the Southern border on May 21. Jackley told KELOLAND News that the next step is for ICE to train the DCI officers to perform under TFM. Jackley said the training requires 40 hours and will be virtual. On May 21, Republican Gov. Larry Rhoden wrote a letter of intent to ICE for Highway Patrol to participate in TFM under 287(g). Rhoden said he's proud of the current work by ICE 'to keep Americans safe and remove illegal alien criminals,' according to the letter of intent, and the Highway Patrol partnership will 'serve and meet the needs of South Dakota, help remove dangerous criminals, and keep drugs out of our communities.' In the letter, Rhoden said five state patrol officers will serve as task force officers. Under 287(g), task force officers enforce immigration authority during their daily duties. Highway patrol duties include traffic law enforcement- like traffic stops- drug interdiction, and more, according to the South Dakota Department of Public Safety website. Rhoden wrote the letter of intent the same day he visited South Dakota National Guard soldiers at the Southern Border. The next step is for Highway Patrol to sign an MOA under the TFM, and submit the letter of intent and MOA to DHS. KELOLAND News reached out to Highway Patrol for a response, but the agency said they were unable to speak this week. As of March, the Hughes and Minnehaha County Sheriff's Offices participates in the 287(g) program as WSOs. 'Recently, with the changes that the federal government has made with the uptake in detentions that we're doing, we just want to be good partners with ICE and our federal agency partners and make sure that we're getting the training that we need to be effective in what they're asking us to do,' Captain Adam Zishka with the Minnehaha County Sheriff's Office said to KELOLAND News in April. As of May 23, Zishka said no officers have been trained yet, and the county continues to wait for details on how many jail officers will receive the Warrant Service Officer training. Patrick Callahan, sheriff for the Hughes County Sheriff's Office, told KELOLAND News that they haven't received training either, as of May 23. 'My intent is to take the training myself and act as the Warrant Service Office at the Hughes County Jail,' Patrick Callahan, sheriff for the Hughes County Sheriff's Office, said in a statement to KELOLAND News in April. 'HCJ is one of the largest jails in South Dakota and holds inmates from across Central South Dakota.' To learn more about the sheriff's offices involvement, read the full article here. Copyright 2025 Nexstar Media, Inc. All rights reserved. This material may not be published, broadcast, rewritten, or redistributed.
