Wei Ji

The Segment Anything Model (SAM) has recently gained popularity in the field of image segmentation due to its impressive capabilities in various segmentation tasks and its prompt-based interface. However, recent studies and individual experiments have shown that SAM underperforms in medical image segmentation due to the lack of medical-specific knowledge. This raises the question of how to enhance SAM's segmentation capability for medical images. We propose the Medical SAM Adapter (Med-SA), which is one of the first methods to integrate SAM into medical image segmentation. Med-SA uses a light yet effective adaptation technique instead of fine-tuning the SAM model, incorporating domain-specific medical knowledge into the segmentation model. We also propose Space-Depth Transpose (SD-Trans) to adapt 2D SAM to 3D medical images and Hyper-Prompting Adapter (HyP-Adpt) to achieve prompt-conditioned adaptation. Comprehensive evaluation experiments on 17 medical image segmentation tasks across various modalities demonstrate the superior performance of Med-SA while updating only 2% of the SAM parameters (13M). Our code is released at https://github.com/KidsWithTokens/Medical-SAM-Adapter.

The Diffusion Probabilistic Model (DPM) has recently gained popularity in the field of computer vision, thanks to its image generation applications, such as Imagen, Latent Diffusion Models, and Stable Diffusion, which have demonstrated impressive capabilities and sparked much discussion within the community. Recent investigations have further unveiled the utility of DPM in the domain of medical image analysis, as underscored by the commendable performance exhibited by the medical image segmentation model across various tasks. Although these models were originally underpinned by a UNet architecture, there exists a potential avenue for enhancing their performance through the integration of vision transformer mechanisms. However, we discovered that simply combining these two models resulted in subpar performance. To effectively integrate these two cutting-edge techniques for the Medical image segmentation, we propose a novel Transformer-based Diffusion framework, called MedSegDiff-V2. We verify its effectiveness on 20 medical image segmentation tasks with different image modalities. Through comprehensive evaluation, our approach demonstrates superiority over prior state-of-the-art (SOTA) methodologies. Code is released at https://github.com/KidsWithTokens/MedSegDiff.

Control over the handedness of circularly polarized luminescence (CPL) in supramolecular gels is of special significance in biology and optoelectronics; however, it still remains a great challenge to precisely and efficiently regulate the chirality of CPL. Herein, a chiral phenylalanine-derived hydrogelator and achiral coumarin derivatives can co-assemble into nanofibrous hydrogels with controllable chirality, and the handedness of CPL of these hydrogels can be efficiently inverted by coumarin derivatives through noncovalent interactions, which can be further tuned at will by incorporating metal ions into the co-assembly. The hydrogen bonds, coordination interactions, and steric hindrance are proved to be the crucial factors for the CPL inversion. This study provides feasible strategies to efficiently regulate the handedness of CPL through co-assembly, and these CPL materials may have potential applications in the fields of photoelectric devices, smart chiroptical materials, and biological systems.

Abstract Realization of a self-assembled, nontoxic and eco-friendly piezoelectric device with high-performance, sensitivity and reliability is highly desirable to complement conventional inorganic and polymer based materials. Hierarchically organized natural materials such as collagen have long been posited to exhibit electromechanical properties that could potentially be amplified via molecular engineering to produce technologically relevant piezoelectricity. Here, by using a simple, minimalistic, building block of collagen, we fabricate a peptide-based piezoelectric generator utilising a radically different helical arrangement of Phe-Phe-derived peptide, Pro-Phe-Phe and Hyp-Phe-Phe, based only on proteinogenic amino acids. The simple addition of a hydroxyl group increases the expected piezoelectric response by an order of magnitude ( d 35 = 27 pm V −1 ). The value is highest predicted to date in short natural peptides. We demonstrate tripeptide-based power generator that produces stable max current >50 nA and potential >1.2 V. Our results provide a promising device demonstration of computationally-guided molecular engineering of piezoelectricity in peptide nanotechnology.