Friday, July 25, 2025

POSTMODERN CAPITALI$M

The secret to success for a unicorn: simple technologies and private funding at an early stage


A new international study led by Kaunas University of Technology, Lithuania reveals that digital mental health start-ups reaching unicorn status tend to secure private funding early, prioritise simple, scalable services, and use subscription models.



Kaunas University of Technology

Prof. Asta Pundzienė, KTU 

image: 

Prof. Asta Pundzienė, Head of the Digital Healthcare Innovation Research Group at KTU

view more 

Credit: KTU





A new study by researchers from Kaunas University of Technology in Lithuania and their partners reveals that digital mental health platform start-ups that have become unicorns often received funding in the early stages, developed necessary services rather than advanced technologies, and used subscriptions as a revenue model. However, according to the researchers, state funding can have the opposite effect, hindering a start-up’s development rather than encouraging it to become a unicorn.

In venture capital terminology, a “unicorn” is a start-up company valued at over 1 billion US dollars. Digital platforms have the potential to become unicorns more quickly than other types of ventures due to their business model. However, according to Asta Pundzienė, a professor at the School of Economics and Business at KTU, fewer than 10% of digital service start-ups achieve such exceptional success.

“In Lithuania and Europe in general, investors tend to prioritise start-ups that develop deep tech,” says Pundzienė, Head of the Digital Healthcare Innovation Research Group at KTU. “However, our research has shown that not only deep tech companies, but also those developing simpler technologies (known as shallow tech) can also achieve exceptional success.”

According to her, the results of the KTU study provide grounds for considering more targeted funding criteria for start-ups, and the insights could be useful for all digital service developers, not only in the health sector.

Success is determined by a combination of factors

A team of researchers from KTU School of Economics and Business, together with colleagues from universities in France and Spain, examined 125 mental health platform ventures, including 12 unicorns, to identify the success factors. According to the researchers, the demand for digital platforms offering mental health services the number of such solutions has grown significantly in the post-COVID period.

“While researching the digital health technologies field, we were intrigued – as to why so often, the ventures which are developing health services, essential for people, are unsuccessful. This question led to another: What are the successful ventures doing differently? The mental health platform sector was chosen as it provided us a sufficiently broad sample for investigating these questions,” says Rima Sermontytė-Baniulė, one of the co-authors of the study.

Summarising the results of the study, Sermontytė-Baniulė states that exceptional success is not determined by a single factor, but by their combination and complementary solutions. After analysing the digital mental health platform unicorns, the team of researchers identified several potentially successful combinations of managerial choices.

“Our research shows that obtaining funding at an early stage of business development is one of the key factors for success. Successful business decisions include the development of deep technologies in the business-to-business segment, not opting for state funding, and the development of services based on shallow tech in the business-to-customer segment, as well as choosing a subscription as a revenue model,” explains the KTU researcher.

According to Prof. Pundzienė, although the study focuses on digital health services, its results could be applied in other similar areas, such as education. The subscription revenue model is also used by digital service companies in other sectors, such as media, finance and information technology.

State funding turns out to be a double-edged sword

One of the most interesting findings of the study is the insight that government funding does not help and, in many cases, even hinders a venture from achieving a unicorn status.

“Governmental institutions that manage venture funds often have different objectives than private investors. For example, the state-managed institutions often pursue social goals, such as employing more researchers, set very high standards for scientific research, and prioritise the development of advanced technologies,” says Pundzienė.

According to her, these goals do not necessarily coincide with the principles of rapid business development. Sermontytė-Baniulė adds that accountability for state funding is based on formal criteria set by state institutions. The state-funded ventures need to focus on meeting these criteria, which are often not compatible with business logic.

Although the study conducted by the KTU Digital Healthcare Innovation Research Group was global in scope, the vast majority of the 125 mental health platforms analysed and all 12 unicorns were based in the US. According to Prof. Pundzienė, these results are not surprising, as 6 of the 10 most valuable unicorns in the world are from the United States.

“The innovation ecosystem in the US is highly developed, with a huge network of private investors. Private businesses invest rationally, funding the ventures that bring value to customers, that develop services customers are willing to pay for. Often, these ventures develop deep tech only when the company is already up and running. Of course, deep tech can provide a competitive advantage in an innovation-saturated market,” says Pundzienė.

According to her, when a start-up is forced to survive from grant to grant and in the innovation ecosystem, there are few private investors, the pace of innovation slows down.

Call to investors: Focus on market needs rather than deep tech

KTU researchers are convinced that the criteria used by the European investment funds should be reviewed. According to Pundzienė, similar issues were highlighted in the Draghi report, recently published by the European Commission.

“Europe is asking why, having an exceptionally high level of fundamental research, we are slower to bring innovations to market than the United States or China. On one hand, Europe has a different social structure, a different work ethic, and different market regulations – that is understandable and perfectly fine. On the other hand, when introducing innovations to the market, Americans first look at what consumers and businesses need, rather than prioritising the scientific novelty of a venture. We could learn from this,” says Prof. Pundzienė.

She believes that when financing ventures, especially those creating innovations in such an important field as healthcare, priority should be given not to how much the idea meets the criteria for evaluating scientific innovation activities set out in the Frascati Manual (2015), but to how relevant the technological service is, i.e. what benefits it brings to the user.

“In Lithuania, we already have e-health, which allows health data to be collected and shared nationwide. However, there are many digital health services that are underdeveloped, such as telehealth, digital therapy, clinical decision support, health service management systems and others. If such services do not exist, then this is an innovation that is worth supporting,” says Pundzienė.

The KTU researchers point out that, according to their research, early-stage funding is a particularly important factor in a venture becoming a unicorn. Therefore, they urge the expansion of the state funding model to include more criteria, such as assessing the benefits a company creates for consumers, the stage of development of the start-up, and the growth needs that the financial injection will serve.

“First, we should see whether the venture meets market demand and, if so, support it to develop its innovation. It is not necessary to start with complicated technology, especially when it requires large and long-term investments. The innovation will enable the company to generate revenue, which will allow it to further develop its innovation by implementing deep techwhich will benefit both the healthcare sector and patients,” says Pundzienė.

This paper is part of a project that has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement. Grant Number: 952012

 

New AI model brings breakthroughs in five-day regional weather forecasting



Institute of Atmospheric Physics, Chinese Academy of Sciences
Diagram illustrating the process of cascaded prediction 

image: 

Diagram illustrating the process of cascaded prediction

view more 

Credit: Congqi Cao




For decades, medium-range weather forecasting—predicting conditions 1 to 5 days ahead—has relied heavily on traditional numerical models. However, this approach often struggles when applied to specific regions with limited historical data.

Researchers at Northwestern Polytechnical University in China have now proposed a novel deep learning–based framework that dramatically improves the accuracy of forecasts, even when data are limited.

To address key challenges in regional forecasting, the team introduced a new method that integrates three major innovations: the use of semantic segmentation models originally designed for medical image analysis; a learnable Gaussian noise mechanism that improves the model’s robustness; and a cascade prediction strategy that breaks the forecasting task into manageable stages. The study is published in Atmospheric and Oceanic Science Letters recently.

“Our goal was to make regional forecasting smarter, faster, and more reliable, even in data-limited scenarios.” says Associate Professor Congqi Cao, corresponding author of the study. “This is especially valuable for areas where a dense network of meteorological observations is not available.”

The method was tested on the East China Regional AI Medium Range Weather Forecasting Competition dataset, which includes 10 years of reanalysis data from ERA5. The task involved using past atmospheric variables to predict five key surface weather indicators—including temperature, wind, and precipitation—every 6 hours for the next 5 days.

The results speak for themselves: the model achieved significant improvements in prediction performance, outperforming many mainstream global AI forecasting models. Specifically, the method reduced temperature forecast errors by 9.3%, improved the precipitation F1-score by 6.8%, and lowered wind speed errors by 12.5%.

“This is the first time semantic segmentation and learnable noise mechanisms have been used together for regional weather forecasting,” explains Prof. Cao. “It opens up new possibilities for accurate forecasting in other data-scarce regions.”

Looking ahead, the team plans to extend their method to real-time systems and apply it to more regions across China. They hope their work will eventually serve public safety, agriculture, and disaster prevention needs—delivering smarter, faster, and more local forecasts when they matter most.

 

 

Ketamine could treat depression by interacting with the brain’s ‘opioid system’



King's College London




Ketamine is a highly effective, fast-acting antidepressant that works even for patients who have not responded to other medications. However, the brain mechanisms important for these rapid treatment effects are yet to be determined.

Researchers at King’s College London, who are investigating why ketamine could be a good treatment for some people with depression, have discovered that the drug’s antidepressant effects involve the brain’s opioid system.

The study, led by King’s College London and published in Nature Medicine, included 26 individuals with clinically diagnosed depression who were given a low dose ketamine infusion across two sessions during neuroimaging.

Before receiving the ketamine infusion, in one session they were given naltrexone, which blocks the opioid receptors in the brain, and in the other they were given a placebo.

Participants were monitored during the infusion in a brain scanner using a method called magnetic resonance spectroscopy (MRS). MRS measured dynamic changes in a brain chemical called glutamate. Depressive symptoms were then assessed using the clinician-rated Montgomery-Åsberg Depression Rating Scale (MADRS) 24-hours after infusion, when ketamine’s antidepressive symptoms peak.

They found that blocking the opioid system reduced both the brain’s glutamatergic response and the antidepressant effects observed the following day, suggesting that the opioid system plays a key role in mediating the antidepressant response.

The study also identified a sex-related effect: the effect of naltrexone on glutamatergic activity appeared more pronounced in males with depression than in the females with depression.

These insights into how ketamine works for different people is essential to personalising treatments.

Dr Luke Jelen, lead author of the study and a Clinical Lecturer in Psychiatry at King’s College London, said: “Ketamine often makes the news for negative reasons. However, at a low dose, ketamine shows enormous potential to offer relief from the symptoms of depression.”

“Understanding whether the opioid system is involved ketamine’s antidepressant effects is a really important question, given how much we still don't know about how ketamine works. “Our study shows that the opioid system is involved and offers insight into how it contributes to ketamine’s effects.”

The authors are keen to highlight that ketamine is not classified as an opioid and does not bind to opioid receptors with high affinity like morphine or heroin. Instead, the findings point to a dynamic interplay between the glutamatergic and opioid systems, which may work together to support ketamine’s rapid antidepressant effects.

Opiates can offer relief from the systems of depression however they are highly addictive. Understanding if and how the opioid system is involved in the effects of ketamine is important to understand why ketamine works and develop new, alternative treatments.

Low-dose ketamine is currently being used to treat depression in private clinics and a small number of NHS clinics. At higher doses it is also used in medicinal anaesthesia. However, it is also used recreationally and if misused can cause serious health problems including irreversible damage to the bladder and kidneys.

Professor Mitul Mehta, a professor of neuroimaging & psychopharmacology at King’s College London, said: “The brain’s different neurochemical systems work together to produce our experiences and behaviour so it is no surprise that the opiate system may have a role in ketamine’s antidepressant effect.”

“We need these kinds of studies to understand exactly what the important brain mechanisms are for antidepressant effects. Understanding more about how ketamine works can lead to treatment being personalised for different people, which is vital for creating safe and effective treatments."

ENDS

Notes to editor: 

  • If you would like to speak to the researchers involved in this study, please email Jo Dungate at the King’s College London press office at joanna.dungate@kcl.ac.uk.

About King’s College London 

King’s College London is amongst the top 40 universities in the world and 5th best in the UK (QS World University Rankings 2026), and one of England’s oldest and most prestigious universities.  With an outstanding reputation for world-class teaching and cutting-edge research, King’s maintained its sixth position for ‘research power’ in the UK (2021 Research Excellence Framework).     

King's has more than 33,000 students (including more than 12,800 postgraduates) from some 150 countries worldwide, and 8,500 staff.   

For nearly 200 years, King’s students and staff have used their knowledge and insight to make a positive impact on people, society and the planet. Focused on delivering positive change at home in London, across the UK and around the world, King’s is building on its history of addressing the world’s most urgent challenges head on to accelerate progress, make discoveries and pioneer innovation. Visit the website to find out more about Vision 2029, which sets out bold ambitions for the future of King’s as we look towards our 200th anniversary.  

World-changing ideas. Life-changing impact: kcl.ac.uk/news