Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Article
  • Published:

Reconfigurable mixed-kernel heterojunction transistors for personalized support vector machine classification


Advances in algorithms and low-power computing hardware imply that machine learning is of potential use in off-grid medical data classification and diagnosis applications such as electrocardiogram interpretation. However, although support vector machine algorithms for electrocardiogram classification show high classification accuracy, hardware implementations for edge applications are impractical due to the complexity and substantial power consumption needed for kernel optimization when using conventional complementary metal–oxide–semiconductor circuits. Here we report reconfigurable mixed-kernel transistors based on dual-gated van der Waals heterojunctions that can generate fully tunable individual and mixed Gaussian and sigmoid functions for analogue support vector machine kernel applications. We show that the heterojunction-generated kernels can be used for arrhythmia detection from electrocardiogram signals with high classification accuracy compared with standard radial basis function kernels. The reconfigurable nature of mixed-kernel heterojunction transistors also allows for personalized detection using Bayesian optimization. A single mixed-kernel heterojunction device can generate the equivalent transfer function of a complementary metal–oxide–semiconductor circuit comprising dozens of transistors and thus provides a low-power approach for support vector machine classification applications.

This is a preview of subscription content, access via your institution

Access options

Buy this article

Prices may be subject to local taxes which are calculated during checkout

Fig. 1: MKH transistor schematic, structure and performance.
Fig. 2: Arrhythmia detection using single-device-generated mixed kernels.
Fig. 3: BO of personalized mixed kernels.
Fig. 4: Mixed-kernel circuit complexity comparison with conventional CMOS.

Similar content being viewed by others

Data availability

The data for all the figures in this Article are available via the Harvard Dataverse repository at

Code availability

The code used in this study is available via GitHub at


  1. Noble, W. S. What is a support vector machine? Nat. Biotechnol. 24, 1565–1567 (2006).

    Article  Google Scholar 

  2. Cervantes, J., Garcia-Lamont, F., Rodríguez-Mazahua, L. & Lopez, A. A comprehensive survey on support vector machine classification: applications, challenges and trends. Neurocomputing 408, 189–215 (2020).

    Article  Google Scholar 

  3. Genov, R. & Cauwenberghs, G. Kerneltron: support vector ‘machine’ in silicon. IEEE Trans. Neural Netw. 14, 1426–1434 (2003).

    Article  MATH  Google Scholar 

  4. Devikanniga, D., Ramu, A. & Haldorai, A. Efficient diagnosis of liver disease using support vector machine optimized with crows search algorithm. EAI Endorsed Trans. Energy Web 7, e10 (2020).

    Google Scholar 

  5. Wang, H., Zheng, B., Yoon, S. W. & Ko, H. S. A support vector machine-based ensemble algorithm for breast cancer diagnosis. Eur. J. Oper. Res. 267, 687–699 (2018).

    Article  MathSciNet  MATH  Google Scholar 

  6. Ahmad, I., Basheri, M., Iqbal, M. J. & Rahim, A. Performance comparison of support vector machine, random forest, and extreme learning machine for intrusion detection. IEEE Access 6, 33789–33795 (2018).

    Article  Google Scholar 

  7. Afifi, S. M., GholamHosseini, H. & Sinha, R. Hardware implementations of SVM on FPGA: a state-of-the-art review of current practice. Int. J. Innov. Sci. Eng. Technol. 2, 733–752 (2015).

    Google Scholar 

  8. Shoeb, A. H. & Guttag, J. V. Application of machine learning to epileptic seizure detection. In Proc. 27th International Conference on Machine Learning (ICML-10) 975–982 (PMLR, 2010).

  9. Bin Altaf, M. A. & Yoo, J. A 1.83 µJ/classification, 8-channel, patient-specific epileptic seizure classification SoC using a non-linear support vector machine. IEEE Trans. Biomed. Circuits Syst. 10, 49–60 (2016).

    Article  Google Scholar 

  10. Kang, K. & Shibata, T. An on-chip-trainable Gaussian-kernel analog support vector machine. IEEE Trans. Circuits Syst. I, Reg. Papers 57, 1513–1524 (2010).

    Article  MathSciNet  Google Scholar 

  11. Zhang, R. & Shibata, T. Fully parallel self-learning analog support vector machine employing compact Gaussian generation circuits. Jpn. J. Appl. Phys. 51, 04DE10 (2012).

    Article  Google Scholar 

  12. Alimisis, V., Gourdouparis, M., Dimas, C. & Sotiriadis, P. P. A 0.6 V, 3.3 nW, adjustable Gaussian crcuit for tunable kernel functions. In 2021 34th SBC/SBMicro/IEEE/ACM Symposium on Integrated Circuits and Systems Design (SBCCI) 1–6 (IEEE, 2021).

  13. Vrtaric, D., Ceperic, V. & Baric, A. Area-efficient differential Gaussian circuit for dedicated hardware implementations of Gaussian function based machine learning algorithms. Neurocomputing 118, 329–333 (2013).

    Article  Google Scholar 

  14. Reda Mohamed, A., Qi, L., Li, Y. & Wang, G. A generic nano-watt power fully tunable 1-D Gaussian kernel circuit for artificial neural network. IEEE Trans. Circuits Syst., II, Exp. Briefs 67, 1529–1533 (2020).

    Google Scholar 

  15. Sangwan, V. K. & Hersam, M. C. Neuromorphic nanoelectronic materials. Nat. Nanotechnol. 15, 517–528 (2020).

    Article  Google Scholar 

  16. Sangwan, V. K. et al. Self-aligned van der Waals heterojunction diodes and transistors. Nano Lett. 18, 1421–1427 (2018).

    Article  Google Scholar 

  17. Sebastian, A., Pannone, A., Subbulakshmi Radhakrishnan, S. & Das, S. Gaussian synapses for probabilistic neural networks. Nat. Commun. 10, 4199 (2019).

    Article  Google Scholar 

  18. Beck, M. E. et al. Spiking neurons from tunable Gaussian heterojunction transistors. Nat. Commun. 11, 1565 (2020).

    Article  Google Scholar 

  19. Duong, D. L., Lee, S. M. & Lee, Y. H. Origin of unipolarity in carbon nanotube field effect transistors. J. Mater. Chem. 22, 1994–1997 (2012).

    Article  Google Scholar 

  20. Wang, K.-C. et al. Atomic-level charge transport mechanism in gate-tunable anti-ambipolar van der Waals heterojunctions. Appl. Phys. Lett. 118, 083103 (2021).

    Article  Google Scholar 

  21. Kim, C. H., Hayakawa, R. & Wakayama, Y. Fundamentals of organic anti‐ambipolar ternary inverters. Adv. Electron. Mater. 6, 1901200 (2020).

    Article  Google Scholar 

  22. Wu, E. et al. Photoinduced doping to enable tunable and high-performance anti-ambipolar MoTe2/MoS2 heterotransistors. ACS Nano 13, 5430–5438 (2019).

    Article  Google Scholar 

  23. Kobashi, K., Hayakawa, R., Chikyow, T. & Wakayama, Y. Multi-valued logic circuits based on organic anti-ambipolar transistors. Nano Lett. 18, 4355–4359 (2018).

    Article  Google Scholar 

  24. Li, Y. et al. Anti-ambipolar field-effect transistors based on few-layer 2D transition metal dichalcogenides. ACS Appl. Mater. Interfaces 8, 15574–15581 (2016).

    Article  Google Scholar 

  25. Song, M. H., Lee, J., Cho, S. P., Lee, K. J. & Yoo, S. K. Support vector machine based arrhythmia classification using reduced features. Int. J. Control Autom. Syst. 3, 571–579 (2005).

    Google Scholar 

  26. Asl, B. M., Setarehdan, S. K. & Mohebbi, M. Support vector machine-based arrhythmia classification using reduced features of heart rate variability signal. Artif. Intell. Med. 44, 51–64 (2008).

    Article  Google Scholar 

  27. Moody, G. B. & Mark, R. G. The impact of the MIT-BIH arrhythmia database. IEEE Eng. Med. Biol. Mag. 20, 45–50 (2001).

    Article  Google Scholar 

  28. Goldberger, A. L. et al. PhysioBank, PhysioToolkit, and PhysioNet: components of a new research resource for complex phiysologic signals. Circulation 101, e215–e220 (2000).

    Article  Google Scholar 

  29. Shahriari, B., Swersky, K., Wang, Z., Adams, R. P. & De Freitas, N. Taking the human out of the loop: a review of Bayesian optimization. Proc. IEEE 104, 148–175 (2015).

    Article  Google Scholar 

  30. Delbruck, T. ‘Bump’ circuits for computing similarity and dissimilarity of analog voltages. In IJCNN-91-Seattle International Joint Conference on Neural Networks 475–479 (IEEE, 1993).

  31. Verleysen, M., Thissen, P., Voz, J.-L. & Madrenas, J. An analog processor architecture for a neural network classifier. IEEE Micro 14, 16–28 (1994).

    Article  Google Scholar 

  32. Nam, M. & Cho, K. Implementation of real-time image edge detector based on a bump circuit and active pixels in a CMOS image sensor. Integration 60, 56–62 (2018).

    Article  Google Scholar 

  33. Payvand, M. & Indiveri, G. Spike-based plasticity circuits for always-on on-line learning in neuromorphic systems. In 2019 IEEE International Symposium on Circuits and Systems (ISCAS) 1–5 (IEEE, 2019).

  34. Lu, J., Yang, T., Jahan, M. & Holleman, J. Nano‐power tunable bump circuit using wide‐input‐range pseudo‐differential transconductor. Electron. Lett. 50, 921–923 (2014).

    Article  Google Scholar 

  35. Alimisis, V., Gourdouparis, M., Gennis, G., Dimas, C. & Sotiriadis, P. P. Analog Gaussian function circuit: architectures, operating principles and applications. Electronics 10, 2530 (2021).

    Article  Google Scholar 

  36. Youssefi, B., Leigh, A. J., Mirhassani, M. & Wu, J. Tunable neuron with PWL approximation based on the minimum operator. IEEE Trans. Circuits Syst., II, Exp. Briefs 66, 387–391 (2018).

    Google Scholar 

  37. Yan, X., Qian, J. H., Sangwan, V. K. & Hersam, M. C. Progress and challenges for memtransistors in neuromorphic circuits and systems. Adv. Mater. 34, 2108025 (2022).

    Article  Google Scholar 

Download references


Device fabrication was supported by the DOE ASCR and BES Microelectronics Threadwork Program, which is funded by the US Department of Energy (DOE), Office of Science, under contract no. DE-AC02-06CH11357. Device testing was supported by the National Science Foundation Materials Research Science and Engineering Center at Northwestern University under contract no. DMR-1720139 and by the National Science Foundation Neuroplane Program under contract no. CCF-2106964. SVM analysis and BO were supported by the Army Research Office under contract no. W911NF-21-2-0128. Materials growth and characterization were supported by the Laboratory Directed Research and Development Program at Sandia National Laboratories (SNL). SNL is a multi-mission laboratory managed and operated by National Technology and Engineering Solutions of Sandia LLC, a wholly owned subsidiary of Honeywell International Inc. for the US DOE National Nuclear Security Administration under contract DE-NA0003525. This work describes objective technical results and analysis. Any subjective views or opinions that might be expressed in the paper do not necessarily represent the views of the United States Government. This work made use of the Northwestern University Micro/Nano Fabrication Facility (NUFAB), which has received support from the Soft and Hybrid Nanotechnology Experimental (SHyNE) Resource (NSF ECCS-1542205), the Materials Research Science and Engineering Center (NSF DMR-1720139) and the State of Illinois.

Author information

Authors and Affiliations



M.C.H., H.W. and V.K.S. conceived the idea. X.Y. and J.H.Q. designed the experiments and performed the device fabrication, measurements and analysis. J.M. performed the SVM analysis and BO. A.Z. and X.W. assisted with the circuit design, analysis and comparison. S.E.L., M.P.B. and K.J.L. synthesized and characterized the MoS2 films. All authors discussed the results and contributed to the writing of the manuscript.

Corresponding authors

Correspondence to Vinod K. Sangwan, Han Wang or Mark C. Hersam.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Electronics thanks Kyusang Lee and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Supplementary Figs. 1–14, Tables 1–3 and Notes 1 and 2.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Yan, X., Qian, J.H., Ma, J. et al. Reconfigurable mixed-kernel heterojunction transistors for personalized support vector machine classification. Nat Electron 6, 862–869 (2023).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI:

This article is cited by


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing