Aurelia Butler

Researchers from Children’s Hospital of Philadelphia (CHOP) have developed a new AI-powered algorithm to help understand how different cells organize themselves into particular tissues and communicate with one another. This new tool was tested on two types of cancer tissues to reveal how these “neighborhoods” of cells interact with one another to evade therapy, and more studies could reveal more information about the function of these cells in the tumor microenvironment.
The findings were published online today by the journal Nature Methods.
To understand how different cells organize themselves to support the functions of a tissue, researchers proposed the concept of tissue cellular neighborhoods (TCNs) to describe functional units in which different, recurrent cell types work together to support specific tissue functions. Across individuals, the functions of these TCNs would remain the same. However, translating the huge amount of information in spatial omics data into models and hypotheses that can be interpreted and tested by researchers requires advanced AI algorithms.
“It is very difficult to study the tissue microenvironment, how certain cells organize, behave and communicate with one another,” said senior study author Kai Tan, PhD, an investigator in the Center for Childhood Cancer Research at CHOP and a professor in the Department of Pediatrics and the Perelman School of Medicine at the University of Pennsylvania. “Until recent advances in so-called spatial omics technology, it was impossible to spatially characterize more than 100 proteins or hundreds or even thousands of genes across a piece of tissue, which might be home to hundreds of thousands of cells and their respective genes.”
In this study, researchers developed the deep-learning-based CytoCommunity algorithm to identify TCNs based on cell identities of a tissue sample, their spatial distributions as well as patient clinical data, which can help researchers better understand how these neighborhoods of cells are organized and are associated with certain clinical outcomes. In this study, tissue samples from breast and colorectal tumors were used because of a high volume of data available, enough to train the algorithm to identify TCNs associated with high-risk disease subtypes.
By using CytoCommunity for breast and colorectal cancer data, the algorithm revealed new fibroblast-enriched TCNs and granulocyte-enriched TCNs specific to high-risk breast cancer and colorectal cancer, respectively.
“Since we were able to prove the effectiveness of CytoCommunity, the next step is to apply this algorithm to both healthy and diseased tissue data generated by research consortia such as HuBMAP (Human BioMolecular Atlas Program) and HTAN (Human Tumor Atlas Network)” Tan said. “For instance, using data from childhood cancers such as leukemia, neuroblastoma and high-grade gliomas, we hope to find tissue cellular neighborhoods that might be associated with responses to certain therapies and combine our findings with genetic data to help determine which genetic pathways may be involved at the cellular and molecular levels.”
This study was supported by National Institutes of Health grant CA233285, HL165442 and HL156090, a grant from the Chan Zuckerberg Initiative (AWD-2021-237920), a grant from the Leona M. and Harry B Helmsley Charitable Trust (no. 2008-04062), a National Natural Science Foundation of China grant no. 62002277, a grant from the Young Talent Fund of University Association for Science and Technology in Shaanxi (no. 20210101), a grant from the Fundamental Research Funds for the Central Universities (no. QTZX23051), and National Natural Science Foundation of China grant nos. 62132015 and U22A2037. More

Freezing is one of the most common and debilitating symptoms of Parkinson’s disease, a neurodegenerative disorder that affects more than 9 million people worldwide. When individuals with Parkinson’s disease freeze, they suddenly lose the ability to move their feet, often mid-stride, resulting in a series of staccato stutter steps that get shorter until the person stops altogether. These episodes are one of the biggest contributors to falls among people living with Parkinson’s disease.
Today, freezing is treated with a range of pharmacological, surgical or behavioral therapies, none of which are particularly effective.
What if there was a way to stop freezing altogether?
Researchers from the Harvard John A. Paulson School of Engineering and Applied Sciences (SEAS) and the Boston University Sargent College of Health & Rehabilitation Sciences have used a soft, wearable robot to help a person living with Parkinson’s walk without freezing. The robotic garment, worn around the hips and thighs, gives a gentle push to the hips as the leg swings, helping the patient achieve a longer stride.
The device completely eliminated the participant’s freezing while walking indoors, allowing them to walk faster and further than they could without the garment’s help.
“We found that just a small amount of mechanical assistance from our soft robotic apparel delivered instantaneous effects and consistently improved walking across a range of conditions for the individual in our study,” said Conor Walsh, the Paul A. Maeder Professor of Engineering and Applied Sciences at SEAS and co-corresponding author of the study.
The research demonstrates the potential of soft robotics to treat this frustrating and potentially dangerous symptom of Parkinson’s disease and could allow people living with the disease to regain not only their mobility but their independence.

The research is published in Nature Medicine.
For over a decade, Walsh’s Biodesign Lab at SEAS has been developing assistive and rehabilitative robotic technologies to improve mobility for individuals’ post-stroke and those living with ALS or other diseases that impact mobility. Some of that technology, specifically an exosuit for post-stroke gait retraining, received support from the Wyss Institute for Biologically Inspired Engineering, and was licensed and commercialized by ReWalk Robotics.
In 2022, SEAS and Sargent College received a grant from the Massachusetts Technology Collaborative to support the development and translation of next-generation robotics and wearable technologies. The research is centered at the Move Lab, whose mission is to support advances in human performance enhancement with the collaborative space, funding, R&D infrastructure, and experience necessary to turn promising research into mature technologies that can be translated through collaboration with industry partners.
This research emerged from that partnership.
“Leveraging soft wearable robots to prevent freezing of gait in patients with Parkinson’s required a collaboration between engineers, rehabilitation scientists, physical therapists, biomechanists and apparel designers,” said Walsh, whose team collaborated closely with that of Terry Ellis, Professor and Physical Therapy Department Chair and Director of the Center for Neurorehabilitation at Boston University.
The team spent six months working with a 73-year-old man with Parkinson’s disease, who — despite using both surgical and pharmacologic treatments — endured substantial and incapacitating freezing episodes more than 10 times a day, causing him to fall frequently. These episodes prevented him from walking around his community and forced him to rely on a scooter to get around outside.

In previous research, Walsh and his team leveraged human-in-the-loop optimization to demonstrate that a soft, wearable device could be used to augment hip flexion and assist in swinging the leg forward to provide an efficient approach to reduce energy expenditure during walking in healthy individuals.
Here, the researchers used the same approach but to address freezing. The wearable device uses cable-driven actuators and sensors worn around the waist and thighs. Using motion data collected by the sensors, algorithms estimate the phase of the gait and generate assistive forces in tandem with muscle movement.
The effect was instantaneous. Without any special training, the patient was able to walk without any freezing indoors and with only occasional episodes outdoors. He was also able to walk and talk without freezing, a rarity without the device.
“Our team was really excited to see the impact of the technology on the participant’s walking,” said Jinsoo Kim, former PhD student at SEAS and co-lead author on the study.
During the study visits, the participant told researchers: “The suit helps me take longer steps and when it is not active, I notice I drag my feet much more. It has really helped me, and I feel it is a positive step forward. It could help me to walk longer and maintain the quality of my life.”
“Our study participants who volunteer their time are real partners,” said Walsh. “Because mobility is difficult, it was a real challenge for this individual to even come into the lab, but we benefited so much from his perspective and feedback.”
The device could also be used to better understand the mechanisms of gait freezing, which is poorly understood.
“Because we don’t really understand freezing, we don’t really know why this approach works so well,” said Ellis. “But this work suggests the potential benefits of a ‘bottom-up’ rather than ‘top-down’ solution to treating gait freezing. We see that restoring almost-normal biomechanics alters the peripheral dynamics of gait and may influence the central processing of gait control.”
The research was co-authored by Jinsoo Kim, Franchino Porciuncula, Hee Doo Yang, Nicholas Wendel, Teresa Baker and Andrew Chin. Asa Eckert-Erdheim and Dorothy Orzel also contributed to the design of the technology, as well as Ada Huang, and Sarah Sullivan managed the clinical research. It was supported by the National Science Foundation under grant CMMI-1925085; the National Institutes of Health under grant NIH U01 TR002775; and the Massachusetts Technology Collaborative, Collaborative Research and Development Matching Grant. More

Spintronic devices are electronic devices that utilize the spin of electrons (an intrinsic form of angular momentum possessed by the electron) to achieve high-speed processing and low-cost data storage. In this regard, spin-transfer torque is a key phenomenon that enables ultrafast and low-power spintronic devices. Recently, however, spin-orbit torque (SOT) has emerged as a promising alternative to spin-transfer torque.
Many studies have investigated the origin of SOT, showing that in non-magnetic materials, a phenomenon called the spin Hall effect (SHE) is key to achieving SOT. In these materials, the existence of a “Dirac band” structure, a specific arrangement of electrons in terms of their energy, is important to achieving large SHE. This is because the Dirac band structure contains “hot spots” for the Berry phase, a quantum phase factor responsible for the intrinsic SHE. Thus, materials with suitable Berry phase hot spots are key to engineering the SHE.
In this context, the material tantalum silicide (TaSi2) is of great interest as it has several Dirac points near the Fermi level in its band structure, suitable for practicing Berry phase engineering. To demonstrate this, a team of researchers, led by Associate Professor Pham Nam Hai from the Department of Electrical and Electronic Engineering at Tokyo Institute of Technology (Tokyo Tech), Japan, recently investigated the influence of Dirac band hot spots on the temperature dependence of SHE in TaSi2. “Berry phase monopole engineering is an interesting avenue of research as it can give rise to efficient high-temperature SOT spintronic devices such as the magneto-resistive random-access memory,” explains Dr. Hai about the importance of their study. Their findings were published in the journal Applied Physics Letters.
Through various experiments, the team observed that the SOT efficiency of TaSi2 remained almost unchanged from 62 K to 288 K, which was similar to the behavior of conventional heavy metals. However, upon increasing the temperature further, the SOT efficiency suddenly increased and nearly doubled at 346 K. Moreover, the corresponding SHE also increased in a similar fashion. Notably, this was quite different from the behavior of conventional heavy metals and their alloys. Upon further analysis, the researchers attributed this sudden increase in SHE at high temperatures to Berry phase monopoles.
“These results provide a strategy to enhance the SOT efficiency at high temperatures via Berry phase monopole engineering,” highlights Dr. Hai.
Indeed, their study highlights the potential of Berry phase monopole engineering to effectively use the SHE in non-magnetic materials, and provides a new pathway for the development of high-temperature, ultrafast, and low-power SOT spintronic devices. More

Researchers at the Georgia Institute of Technology have created the world’s first functional semiconductor made from graphene, a single sheet of carbon atoms held together by the strongest bonds known. Semiconductors, which are materials that conduct electricity under specific conditions, are foundational components of electronic devices. The team’s breakthrough throws open the door to a new way of doing electronics.
Their discovery comes at a time when silicon, the material from which nearly all modern electronics are made, is reaching its limit in the face of increasingly faster computing and smaller electronic devices. Walter de Heer, Regents’ Professor of physics at Georgia Tech, led a team of researchers based in Atlanta, Georgia, and Tianjin, China, to produce a graphene semiconductor that is compatible with conventional microelectronics processing methods — a necessity for any viable alternative to silicon.
In this latest research, published in Nature, de Heer and his team overcame the paramount hurdle that has been plaguing graphene research for decades, and the reason why many thought graphene electronics would never work. Known as the “band gap,” it is a crucial electronic property that allows semiconductors to switch on and off. Graphene didn’t have a band gap — until now.
“We now have an extremely robust graphene semiconductor with 10 times the mobility of silicon, and which also has unique properties not available in silicon,” de Heer said. “But the story of our work for the past 10 years has been, ‘Can we get this material to be good enough to work?'”
A New Type of Semiconductor
De Heer started to explore carbon-based materials as potential semiconductors early in his career, and then made the switch to exploring two-dimensional graphene in 2001. He knew then that graphene had potential for electronics.
“We were motivated by the hope of introducing three special properties of graphene into electronics,” he said. “It’s an extremely robust material, one that can handle very large currents, and can do so without heating up and falling apart.”
De Heer achieved a breakthrough when he and his team figured out how to grow graphene on silicon carbide wafers using special furnaces. They produced epitaxial graphene, which is a single layer that grows on a crystal face of the silicon carbide. The team found that when it was made properly, the epitaxial graphene chemically bonded to the silicon carbide and started to show semiconducting properties.

Over the next decade, they persisted in perfecting the material at Georgia Tech and later in collaboration with colleagues at the Tianjin International Center for Nanoparticles and Nanosystems at Tianjin University in China. De Heer founded the center in 2014 with Lei Ma, the center’s director and a co-author of the paper.
How They Did It
In its natural form, graphene is neither a semiconductor nor a metal, but a semimetal. A band gap is a material that can be turned on and off when an electric field is applied to it, which is how all transistors and silicon electronics work. The major question in graphene electronics research was how to switch it on and off so it can work like silicon.
But to make a functional transistor, a semiconducting material must be greatly manipulated, which can damage its properties. To prove that their platform could function as a viable semiconductor, the team needed to measure its electronic properties without damaging it.
They put atoms on the graphene that “donate” electrons to the system — a technique called doping, used to see whether the material was a good conductor. It worked without damaging the material or its properties.
The team’s measurements showed that their graphene semiconductor has 10 times greater mobility than silicon. In other words, the electrons move with very low resistance, which, in electronics, translates to faster computing. “It’s like driving on a gravel road versus driving on a freeway,” de Heer said. “It’s more efficient, it doesn’t heat up as much, and it allows for higher speeds so that the electrons can move faster.”
The team’s product is currently the only two-dimensional semiconductor that has all the necessary properties to be used in nanoelectronics, and its electrical properties are far superior to any other 2D semiconductors currently in development.

“A long-standing problem in graphene electronics is that graphene didn’t have the right band gap and couldn’t switch on and off at the correct ratio,” said Ma. “Over the years, many have tried to address this with a variety of methods. Our technology achieves the band gap, and is a crucial step in realizing graphene-based electronics.”
Moving Forward
Epitaxial graphene could cause a paradigm shift in the field of electronics and allow for completely new technologies that take advantage of its unique properties. The material allows the quantum mechanical wave properties of electrons to be utilized, which is a requirement for quantum computing.
“Our motivation for doing graphene electronics has been there for a long time, and the rest was just making it happen,” de Heer said. “We had to learn how to treat the material, how to make it better and better, and finally how to measure the properties. That took a very, very long time.”
According to de Heer, it is not unusual to see yet another generation of electronics on its way. Before silicon, there were vacuum tubes, and before that, there were wires and telegraphs. Silicon is one of many steps in the history of electronics, and the next step could be graphene.
“To me, this is like a Wright brothers moment,” de Heer said. “They built a plane that could fly 300 feet through the air. But the skeptics asked why the world would need flight when it already had fast trains and boats. But they persisted, and it was the beginning of a technology that can take people across oceans.” More

Coal is an abundant resource in the United States that has, unfortunately, contributed to climate change through its use as a fossil fuel. As the country transitions to other means of energy production, it will be important to consider and reevaluate coal’s economic role. A joint research effort from the University of Illinois Urbana-Champaign, the National Energy Technology Laboratory, Oak Ridge National Laboratory and the Taiwan Semiconductor Manufacturing Company has shown how coal can play a vital role in next-generation electronic devices.
“Coal is usually thought of as something bulky and dirty, but the processing techniques we’ve developed can transform it into high-purity materials just a couple of atoms thick,” said Qing Cao, a U. of I. materials science & engineering professor and a co-lead of the collaboration. “Their unique atomic structures and properties are ideal for making some of the smallest possible electronics with performance superior to state-of-the art.”
A process developed by the NETL first converts coal char into nanoscale carbon disks called “carbon dots” that the U. of I. research group demonstrated can be connected to form atomically thin membranes for applications in both two-dimensional transistors and memristors, technologies that will be critical to constructing more advanced electronics. These results are reported in the journal Communications Engineering.
Perfect for 2D electronics
In the ongoing search for smaller, faster and more efficient electronics, the final step will be devices made with materials just one or two atoms thick. It is impossible for devices to be smaller than this limit, and their small scale often makes them operate much quicker and consume far less energy. While ultrathin semiconductors have been extensively studied, it is also necessary to have atomically thin insulators — materials that block electric currents — to construct working electronic devices like transistors and memristors.
Atomically thin layers of carbon with disordered atomic structures can function as an excellent insulator for constructing two-dimensional devices. The researchers in the collaboration have shown that such carbon layers can be formed from carbon dots derived from coal char. To demonstrate their capabilities, the U. of I. group led by Cao developed two examples of two-dimensional devices.
“It’s really quite exciting, because this is the first time that coal, something we normally see as low-tech, has been directly linked to the cutting edge of microelectronics,” Cao said.

Transistor dielectric
Cao’s group used coal-derived carbon layers as the gate dielectric in two-dimensional transistors built on the semimetal graphene or semiconductor molybdenum disulfide to enable more than two times faster device operating speed with lower energy consumption. Like other atomically thin materials, the coal-derived carbon layers do not possess “dangling bonds,” or electrons that are not associated with a chemical bond. These sites, which are abundant on the surface of conventional three-dimensional insulators, alter their electrical properties by effectively functioning as “traps,” slowing down the transport of mobile charges and thus the transistor switching speed.
However, unlike other atomically thin materials, the new coal-derived carbon layers are amorphous, meaning that they do not possess a regular, crystalline structure. They therefore do not have boundaries between different crystalline regions that serve as conduction pathways leading to “leakage,” where undesired electrical currents flow through the insulator and cause substantial additional power consumption during device operations.
Memristor filament
Another application Cao’s group considered is memristors — electronic components capable of both storing and operating on data to greatly enhance the implementation of AI technology. These devices store and represent data by modulating a conductive filament formed by electrochemical reactions between a pair of electrodes with the insulator sandwiched in between.
The researchers found that the adoption of ultrathin coal-derived carbon layers as the insulator allows the fast formation of such filament with low energy consumption to enable high device operating speed with low power. Moreover, atomic size rings in these coal-derived carbon layers confine the filament to enhance the reproducible device operations for enhanced data storage fidelity and reliability.
From research to production
The new devices developed by the Cao group provide proof-of-principle for the use of coal-derived carbon layers in two-dimensional devices. What remains is to show that such devices can be manufactured on large scales.
“The semiconductor industry, including our collaborators at Taiwan Semiconductor, is very interested in the capabilities of two-dimensional devices, and we’re trying to fulfill that promise,” Cao said. “Over the next few years, the U. of I. will continue to collaborate with NETL to develop a fabrication process for coal-based carbon insulators that can be implemented in industrial settings.” More

ICFO and Qurv researchers have fabricated a new high-performance shortwave infrared (SWIR) image sensor based on non-toxic colloidal quantum dots. In their study published in Nature Photonics, they report on a new method for synthesizing functional high-quality non-toxic colloidal quantum dots integrable with complementary metal-oxide-semiconductor (CMOS) technology.
Invisible to our eyes, shortwave infrared (SWIR) light can enable unprecedented reliability, function and performance in high-volume, computer vision first applications in service robotics, automotive and consumer electronics markets. Image sensors with SWIR sensitivity can operate reliably under adverse conditions such as bright sunlight, fog, haze and smoke. Furthermore, the SWIR range provides eye-safe illumination sources and opens up the possibility of detecting material properties through molecular imaging.
Colloidal quantum dots (CQD) based image sensor technology offers a promising technology platform to enable high-volume compatible image sensors in the SWIR. CQDs, nanometric semiconductor crystals, are a solution-processed material platform that can be integrated with CMOS and enables accessing the SWIR range. However, a fundamental roadblock exists in translating SWIR-sensitive quantum dots into key enabling technology for mass-market applications, as they often contain heavy metals like lead or mercury (IV-VI Pb, Hg-chalcogenide semiconductors). These materials are subject to regulations by the Restriction of Hazardous Substances (RoHS), a European directive that regulates their use in commercial consumer electronic applications.
In a new study published in Nature Photonics, ICFO researchers Yongjie Wang, Lucheng Peng, and Aditya Malla led by ICREA Prof. at ICFO Gerasimos Konstantatos, in collaboration with researchers Julien Schreier, Yu Bi, Andres Black, and Stijn Goossens, from Qurv, have reported on the development of high-performance infrared photodetectors and a shortwave infrared (SWIR) image sensor operating at room temperature based on non-toxic colloidal quantum dots. The study describes a new method for synthesizing size tuneable, phosphine-free silver telluride (Ag2Te) quantum dots while preserving the advantageous properties of traditional heavy-metal counterparts paving the way to the introduction of SWIR colloidal quantum dot technology in high-volume markets.
While investigating how to synthetize silver bismuth telluride (AgBiTe2) nanocrystals to extent the spectral coverage of the AsBiS2 technology to enhance the performance of photovoltaic devices, the researchers obtained silver telluride (Ag2Te) as a by-product. This material showed a strong and tuneable quantum confined absorption akin to quantum dots. They realized its potential for SWIR photodetectors and image sensors and pivoted their efforts to achieve and control a new process to synthesize phosphine-free versions of silver telluride quantum dots, as phosphine was found to have a detrimental impact on the optoelectronic properties of the quantum dots relevant to photodetection.
In their new synthetic method, the team used different phosphine-free complexes such as a tellurium and silver precursors that led them to obtain quantum dots with a well-controlled size distribution and excitonic peaks over a very broad range of the spectrum. After fabricating and characterizing them, the new synthesized quantum dots exhibited remarkable performances, with distinct excitonic peaks over 1500 nm — an unprecedented achievement compared to previous phosphine-based techniques for quantum dot fabrication.
The researchers decided then to implement the obtained phosphine-free quantum dots to fabricate a simple laboratory scale photodetector on the common standard ITO (Indium Tin Oxide)-coated glass substrate to characterize the devices and measure its properties. “Those lab-scale devices are operated with shining light from the bottom. For CMOS integrated CQD stacks, light comes from the top, whereas the bottom part of the device is taken by the CMOS electronics,” comments Yongjie Wang, postdoc researcher at ICFO and first author of the study. “So, the first challenge we had to overcome was reverting the device setup. A process that in theory sounds simple, but in reality proved to be a challenging task.”
Initially, the photodiode exhibited a low performance in sensing SWIR light, prompting a redesign that incorporated a buffer layer. This adjustment significantly enhanced the photodetector performance, resulting in a SWIR photodiode exhibiting a spectral range from 350nm to 1600nm, a linear dynamic range exceeding 118 dB, a -3dB bandwidth surpassing 110 kHz and a room temperature detectivity of the order 1012 Jones.

“To the best of our knowledge, the photodiodes reported here have for the first time realized solution processed, non-toxic shortwave infrared photodiodes with figures of merit on par with other heavy-metal containing counterparts,” Gerasimos Konstantatos, ICREA Prof. at ICFO and leading author of the study mentions. “These results further support the fact that Ag2Te quantum dots emerge as a promising RoHS-compliant material for low-cost, high-performance SWIR photodetectors applications.”
With the successful development of this heavy-metal-free quantum dot based photodetector, the researchers went further and teamed up with Qurv, an ICFO spin-off, to demonstrate its potential by constructing a SWIR image sensor as a case study. The team integrated the new photodiode with a CMOS based read-out integrated circuit (ROIC) focal plane array (FPA) demonstrating for the first time a proof-of-concept, non-toxic, room temperature-operating SWIR quantum dot based image sensor. The authors of the study tested the imager to prove its operation in the SWIR by taking several pictures of a target object. In particular, they were able to image the transmission of silicon wafers under the SWIR light as well as to visualize the content of plastic bottles that were opaque in the visible light range.
“Accessing the SWIR with a low-cost technology for consumer electronics will unleash the potential of this spectral range with a huge range of applications including improved vision systems for automotive industry (cars) enabling vision and driving under adverse weather conditions,” says Gerasimos Konstantatos. “SWIR band around 1.35-1.40 µm, can provide an eye-safe window, free of background light under day/night condition, thus, further enable long-range light detection and ranging (LiDAR), three-dimensional imaging for automotive, augmented reality and virtual reality applications.”
Now the researchers want to increase the performance of photodiodes by engineering the stack of layers that comprise the photodetector device. They also want to explore new surface chemistries for the Ag2Te quantum dots to improve the performance and the thermal and environmental stability of the material on its way to the market. More

A new artificial intelligence tool that interprets medical images with unprecedented clarity does so in a way that could allow time-strapped clinicians to dedicate their attention to critical aspects of disease diagnosis and image interpretation.
The tool, called iStar (Inferring Super-Resolution Tissue Architecture), was developed by researchers at the Perelman School of Medicine at the University of Pennsylvania, who believe they can help clinicians diagnose and better treat cancers that might otherwise go undetected. The imaging technique provides both highly detailed views of individual cells and a broader look of the full spectrum of how people’s genes operate, which would allow doctors and researchers to see cancer cells that might otherwise have been virtually invisible. This tool can be used to determine whether safe margins were achieved through cancer surgeries and automatically provide annotation for microscopic images, paving the way for molecular disease diagnosis at that level.
A paper on the method, led by Daiwei “David” Zhang, PhD, a research associate, and Mingyao Li, PhD, a professor of Biostatistics and Digital Pathology, was published today in Nature Biotechnology.
Li said that iStar has the ability to automatically detect critical anti-tumor immune formations called “tertiary lymphoid structures,” whose presence correlates with a patient’s likely survival and favorable response to immunotherapy, which is often given for cancer and requires high precision in patient selection. This means, Li said, that iStar could be a powerful tool for determining which patients would benefit most from immunotherapy.
The development of iStar was taken on as part of the field of spatial transcriptomics, a relatively new field used to map gene activities within the space of tissues. Li and her colleagues adapted a machine learning tool called the Hierarchical Vision Transformer and trained it on standard tissue images. It begins by breaking down images into different stages, starting small and looking for fine details, then moving up and “grasping broader tissue patterns,” according to Li. A network guided by the AI system within iStar uses the information from the Hierarchical Vision Transformer to then absorb all of that information and apply it to predict gene activities, often at near-single-cell resolution.
“The power of iStar stems from its advanced techniques, which mirror, in reverse, how a pathologist would study a tissue sample,” Li explained. “Just as a pathologist identifies broader regions and then zooms in on detailed cellular structures, iStar can capture the overarching tissue structures and also focus on the minutiae in a tissue image.”
To test the efficacy of the tool, Li and her colleagues evaluated iStar on many different types of cancer tissue, including breast, prostate, kidney, and colorectal cancers, mixed with healthy tissues. Within these tests, iStar was able to automatically detect tumor and cancer cells that were hard to identify just by eye. Clinicians in the future may be able to pick up and diagnose more hard-to-see or hard-to-identify cancers with iStar acting as a layer of support.

In addition to the clinical possibilities presented by the iStar technique, the tool moves extremely quickly compared to other, similar AI tools. For example, when set up with the breast cancer dataset the team used, iStar finished its analysis in just nine minutes. By contrast, the best competitor AI tool took more than 32 hours to come up with a similar analysis.
That means iStar was 213 times faster.
“The implication is that iStar can be applied to a large number of samples, which is critical in large-scale biomedical studies,” Li said. “Its speed is also important for its current extensions in 3D and biobank sample prediction. In the 3D context, a tissue block may involve hundreds to thousands of serially cut tissue slices. The speed of iStar makes it possible to reconstruct this huge amount of spatial data within a short period of time.”
And the same goes for biobanks, which store thousands, if not millions, of samples. This is where Li and her colleagues are next aiming their research and extension of iStar. They hope to help researchers gain better understandings of the microenvironments within tissues, which could provide more data for diagnostic and treatment purposes moving forward.
This research was funded by the National Institutes of Health (R01GM125301 and R01HG013185). More

New Cornell University-led research finds that social media platforms and the metrics that reward content creators for revealing their innermost selves to fans open creators up to identity-based harassment.
“Creators share deeply personal — often vulnerable — elements of their lives with followers and the wider public,” said Brooke Erin Duffy, associate professor of communication. “Such disclosures are a key way that influencers build intimacy with audiences and form communities. There’s a pervasive sense that internet users clamor for less polished, less idealized, more relatable moments — especially since the pandemic.”
Duffy is the lead author of “Influencers, Platforms, and the Politics of Vulnerability” published in the European Journal of Cultural Studies.
The research team conducted in-depth interviews with content creators to get a sense of how they experience the demands to make their content — and often themselves — visible to audiences, sponsors and the platforms.
Among their findings: The value of vulnerability for platform-based influencers cannot be overstated — authenticity sells, and that means projecting intimacies, insecurities and even secrets; These authentic revelations are often tied to one’s identities, which can open a person up to attacks based on gender, race, sexuality and other perceived traits; Personal and social vulnerabilities were often compounded by the vulnerabilities of platform-dependent labor: Not only did participants identify the failures of their platforms to protect them from harm (as “contractors” instead of “employees”), many felt these companies incentivize networked antagonism.”Influencers and creators have relatively few formal sources of support or protection,” Duffy said. “In contrast to those legally employed by Meta, Twitch and TikTok, creators are independent contractors. They’re left wanting for a lot of the workplace protections traditionally afforded to employees.”
The researchers examined informal strategies — both anticipatory and reactive — that creators deploy to manage their vulnerabilities. The former included the use of platform filtering systems to sift out abusive, profane or hurtful language. The latter strategies ranged from simply not reading the comments to employing the platform’s tools to minimize the impact of what, for many, felt like an inevitable onslaught of critique.
The authors acknowledge the difficulties of resolving endemic issues of internet hate and harassment. “‘Getting off the internet’ is hardly a viable option for participants in the put-yourself-out-there neoliberal job economy,” they wrote — and offer a warning to those wishing to join the creator economy.
“It is something of a truism that ‘everyone gets the same platform,'” they wrote. “We would caution, however, that the politics of visibility — and hence, the politics of vulnerability — are far less egalitarian that platforms lead us to believe.” More