Computational pathology: A survey review and the way forward

Computational Pathology (CPath) is an interdisciplinary science that augments developments of computational approaches to analyze and model medical histopathology images. The main objective for CPath is to develop infrastructure and workflows of digital diagnostics as an assistive CAD system for clinical pathology, facilitating transformational changes in the diagnosis and treatment of cancer that are mainly address by CPath tools. With evergrowing developments in deep learning and computer vision algorithms, and the ease of the data flow from digital pathology, currently CPath is witnessing a paradigm shift. Despite the sheer volume of engineering and scientific works being introduced for cancer image analysis, there is still a considerable gap of adopting and integrating these algorithms in clinical practice. This raises a significant question regarding the direction and trends that are undertaken in CPath. In this article we provide a comprehensive review of more than 800 papers to address the challenges faced in problem design all-the-way to the application and implementation viewpoints. We have catalogued each paper into a model-card by examining the key works and challenges faced to layout the current landscape in CPath. We hope this helps the community to locate relevant works and facilitate understanding of the field's future directions. In a nutshell, we oversee the CPath developments in cycle of stages which are required to be cohesively linked together to address the challenges associated with such multidisciplinary science. We overview this cycle from different perspectives of data-centric, model-centric, and application-centric problems. We finally sketch remaining challenges and provide directions for future technical developments and clinical integration of CPath. For updated information on this survey review paper and accessing to the original model cards repository, please refer to GitHub. Updated version of this draft can also be found from arXiv.

Permutationally Invariant Deep Learning Approach to Molecular Fingerprinting with Application to Compound Mixtures

Recent advancements in deep learning have led to widespread applications of its algorithms to synthetic planning and reaction predictions in the field of chemistry. One major area, known as supervised learning, is being explored for predicting certain properties such as reaction yields and types. Many chemical descriptors known as fingerprints are being explored as potential candidates for reaction properties prediction. However, there are few studies that describe the permutational invariance of chemical fingerprints, which are concatenated at some stage before being fed to deep learning architecture. In this work, we show that by utilizing permutational invariance, we consistently see improved results in terms of accuracy relative to previously published studies. Furthermore, we are able to accurately predict hydrogen peroxide loss with our own dataset, which consists of more than 20 ingredients in each chemical formulation.

Ligand-Stabilized Reduced-Dimensionality Perovskites

Metal halide perovskites have rapidly advanced thin-film photovoltaic performance; as a result, the materials' observed instabilities urgently require a solution. Using density functional theory (DFT), we show that a low energy of formation, exacerbated in the presence of humidity, explains the propensity of perovskites to decompose back into their precursors. We find, also using DFT, that intercalation of phenylethylammonium between perovskite layers introduces quantitatively appreciable van der Waals interactions. These drive an increased formation energy and should therefore improve material stability. Here we report reduced-dimensionality (quasi-2D) perovskite films that exhibit improved stability while retaining the high performance of conventional three-dimensional perovskites. Continuous tuning of the dimensionality, as assessed using photophysical studies, is achieved by the choice of stoichiometry in materials synthesis. We achieve the first certified hysteresis-free solar power conversion in a planar perovskite solar cell, obtaining a 15.3% certified PCE, and observe greatly improved performance longevity.

Perovskite-fullerene hybrid materials suppress hysteresis in planar diodes

Solution-processed planar perovskite devices are highly desirable in a wide variety of optoelectronic applications; however, they are prone to hysteresis and current instabilities. Here we report the first perovskite-PCBM hybrid solid with significantly reduced hysteresis and recombination loss achieved in a single step. This new material displays an efficient electrically coupled microstructure: PCBM is homogeneously distributed throughout the film at perovskite grain boundaries. The PCBM passivates the key PbI3(-) antisite defects during the perovskite self-assembly, as revealed by theory and experiment. Photoluminescence transient spectroscopy proves that the PCBM phase promotes electron extraction. We showcase this mixed material in planar solar cells that feature low hysteresis and enhanced photovoltage. Using conductive AFM studies, we reveal the memristive properties of perovskite films. We close by positing that PCBM, by tying up both halide-rich antisites and unincorporated halides, reduces electric field-induced anion migration that may give rise to hysteresis and unstable diode behaviour.

Materials processing routes to trap-free halide perovskites

Photovoltaic devices based on lead iodide perovskite films have seen rapid advancements, recently achieving an impressive 17.9% certified solar power conversion efficiency. Reports have consistently emphasized that the specific choice of growth conditions and chemical precursors is central to achieving superior performance from these materials; yet the roles and mechanisms underlying the selection of materials processing route is poorly understood. Here we show that films grown under iodine-rich conditions are prone to a high density of deep electronic traps (recombination centers), while the use of a chloride precursor avoids the formation of key defects (Pb atom substituted by I) responsible for short diffusion lengths and poor photovoltaic performance. Furthermore, the lowest-energy surfaces of perovskite crystals are found to be entirely trap-free, preserving both electron and hole delocalization to a remarkable degree, helping to account for explaining the success of polycrystalline perovskite films. We construct perovskite films from I-poor conditions using a lead acetate precursor, and our measurement of a long (600 ± 40 nm) diffusion length confirms this new picture of the importance of growth conditions.