Aipore
| Aipore | |
|---|---|
| Industry | Nanopore sensing, machine learning |
| Founded | 2018, Osaka, Japan |
| Fate | Concluded its activities in 2025 |
| Headquarters | Japan |
| Products | AI particle identification system for the nanoSCOUTER |
Aipore was a Japanese company that combined solid-state nanopores with machine learning to identify viruses and other particles. A page of the Taniguchi laboratory at Osaka University describes it as "a venture company that developed a measurement system combining nanopores (nanoscale pores) with AI". It was founded in 2018 and, the page says, "concluded its activities in 2025".[1]
Its machine-learning software ran on the nanoSCOUTER, a desktop particle measuring instrument made by the test-equipment company Advantest.[2] Molecular Reality Corporation cites one of the Osaka papers as a study that gave it confidence to try to build a nanopore system for the COVID-19 pandemic.[3]
Science
A virus passing through a nanopore blocks some of the ionic current, as any particle does in a Coulter counter. The 2018 paper from the Osaka group asked whether the pattern of that current identifies the virus. In a silicon nitride pore, with electroosmotic flow used to filter contaminants at the pore's orifice, the group classified allotypes of influenza virus from the ionic current signatures of single virions with 68 percent accuracy. Pooling the pulses of more than 20 virions raised the accuracy above 95 percent.[4]
In 2021 the group published a platform with three parts: machine-learning software on a server, a portable instrument that measured current at high speed and precision, and semiconductor nanopore modules that could be made in quantity.[5] Advantest's release says that Aipore supplied the identification software and Advantest the nanoSCOUTER.[6] The platform separated four coronaviruses of similar size (HCoV-229E, SARS-CoV, MERS-CoV and SARS-CoV-2) and detected SARS-CoV-2 in saliva with 90 percent sensitivity and 96 percent specificity, in a five-minute measurement. The test used 100 saliva samples, 50 positive and 50 negative.[5][6]
Products and use
On October 23, 2020, the Osaka group announced its first results from the nanoSCOUTER, and on October 28 Advantest issued a release about them. The release describes "a lightweight, desktop-sized fine particle measurement instrument" that relies on "Aipore's machine learning functionality", says Advantest donated two instruments to the university, and adds that "the product is a scientific instrument, not an approved medical device." The release of June 30, 2021, repeated that statement and called Aipore's part "the AI-based particle identification system".[2][6]
The Taniguchi laboratory's page says the system "was commercialized as an analytical instrument and sold on the market" and that "a COVID-19 testing device built on this system was validated using more than 10,000 real clinical samples". It adds that the system "was also used for screening at Japan's National High School Baseball Championship (Spring Koshien)."[1]
In 2024 Advantest launched a new nanoSCOUTER, the WEL2100, which reports the quantity, size and zeta potential of particles such as lipid nanoparticles, viruses and exosomes. Its release describes the sensor as an electrical method based on the Coulter principle and does not mention Aipore.[7]
Ending
The university page gives the year, 2025, and the phrase "concluded its activities", and nothing more. It uses the same words for a second Osaka venture, Quantum Biosystems, which was founded in 2013 and concluded its activities in 2023.[1] A Debye Length newsletter of October 12, 2026, noted that Aipore concluded activities in 2025.[8]
In Molecular Reality's usage
In a post of March 29, 2026, the founder of Molecular Reality wrote that the 2018 influenza paper "became super important to us in January 2020", when the company "saw covid coming" and tried to launch its Molecular Aquarium, and he called it "a key paper that gave us confidence". The Osaka researchers had trained "a very simple (primitive by modern ML standards) machine learning model" on pulses from single virions of different serotypes of one virus, serotypes that are "all roughly the same size and shape", and, he wrote, it worked. He added that they went on to build "a startup called AIpore that actually shipped a working product."[3] The episode is told in CoronaPore.
See also
References
- ^ a b c "Startups". Taniguchi Lab, SANKEN, Osaka University. Retrieved October 25, 2026.
- ^ a b "Advantest's nanoSCOUTER Fine Particle Measurement Instrument Identifies Covid-19 With High Accuracy" (Press release). Advantest. October 28, 2020.
- ^ a b Kemmishtree (March 29, 2026). "(no subject)". EQBio-OS.
- ^ Arima, A. et al. (2018). "Selective detections of single-viruses using solid-state nanopores". Scientific Reports 8: 16305. doi:10.1038/s41598-018-34665-4.
- ^ a b Taniguchi, M. et al. (2021). "Combining machine learning and nanopore construction creates an artificial intelligence nanopore for coronavirus detection". Nature Communications 12: 3726. doi:10.1038/s41467-021-24001-2.
- ^ a b c "Results of Research on Identification of Covid-19 Viruses Utilizing Advantest's nanoSCOUTER" (Press release). Advantest. June 30, 2021.
- ^ "Advantest Launches New nanoSCOUTER Particle Measuring Equipment" (Press release). Advantest. June 17, 2024.
- ^ Ferris Calloway (October 12, 2026). "Notes on demonpore, Two Pore Guys and Roswell". Debye Length.