arXiv subjects

Thomas Bocklitz

Publications and source records attributed to Thomas Bocklitz.

17 recordsLinked to original sources

arXiv2026

iPINN for Broadband CARS Phase Retrieval: A Framework for Function Approximation and Inverse Modeling Problems in Nonlinear Spectroscopy

Phase retrieval in broadband coherent anti-Stokes Raman spectroscopy (BCARS) is an ill-posed inverse problem. The Raman-like signal is encoded in the imaginary part of the resonant susceptibility, which mixes coherently with a non-resonant background (NRB) that varies across acquisitions. We introduce an inverse physics-informed neural network (iPINN) that predicts Lorentzian peak parameters from raw BCARS spectra and reconstructs the resonant susceptibility through a differentiable analytical forward model. A transformer encoder assigns spectral features to 24 learnable peak slots, and a multi-view consistency loss enforces invariance across NRB pattern, NRB strength, and noise. Unlike direct spectral regression approaches, the method retains accuracy under varying acquisition conditions. On a public benchmark, iPINN achieves the lowest error among the tested baselines (MAE 0.016 vs. next-best 0.046). On 28 zero-shot test spectra acquired across seven solvents and four focal positions, accuracy is depth-invariant in five of seven solvents. These results show that inverse parametric prediction with a differentiable physical decoder supports robust phase retrieval across measurement conditions.

cs.LG↗

arXiv2026

Physics-Informed Deep Learning Model for Cross-Modality Super-Resolution in Fluorescence Microscopy

Cross-modality image translation offers a route to super-resolution fluorescence microscopy from low-resolution images while reducing phototoxicity and instrumentation demands. However, purely data-driven models can produce visually plausible outputs that are inconsistent with optical image formation. Here, we propose a physics-informed generative adversarial network for confocal-to-STED image translation that incorporates microscope-specific point spread function information into the training objective. Simulated and experimentally measured PSFs were evaluated using a limited paired confocal-STED dataset of TOM20-labeled mitochondria in human primary M2 macrophages acquired across different experimental days. Performance was assessed using reference-based and non-reference-based image-quality metrics, together with complementary frequency- and distribution-sensitive analyses. The no-reference metrics probed physics-relevant image properties, including spatial-frequency content, contrast, and signal-to-noise behavior. PSF-guided models improved structural fidelity, reduced local deviations, and achieved closer agreement with STED references than non-PSF baselines, particularly in frequency-domain analyses. These results demonstrate that optical priors can improve the structural fidelity and physical plausibility of generative microscopy models for cross-modality super-resolution imaging.

cs.CV↗

arXiv2026

Transformer-Based Classification of Bacterial Raman Spectra with LOOCV

Transformer-based models have recently attracted increasing attention for Raman spectral classification. In this study, a transformer-based approach was systematically evaluated using a nested leave-one-replicate-out cross-validation framework and compared with conventional machine-learning pipelines combining PCA or ICA with LDA, SVM, and Random Forest classifiers. A bacterial Raman dataset comprising 5,417 single-cell spectra from six bacterial species and nine independent measurement replicates was used. The transformer consistently achieved the highest classification performance across independent test replicates and significantly outperformed all conventional approaches. Analysis of the learned latent feature space revealed improved class separation compared with PCA- and ICA-based representations. Furthermore, the transformer maintained superior performance when applied directly to raw Raman spectra without preprocessing, demonstrating robust behavior across measurement replicates. These findings highlight the potential of transformer-based models for robust Raman spectral classification and emphasize the importance of replicate-aware validation for realistic model evaluation.

cs.LG↗

arXiv2026

Lit2Vec: A Reproducible Workflow for Building a Legally Screened Chemistry Corpus from S2ORC for Downstream Retrieval and Text Mining

We present Lit2Vec, a reproducible workflow for constructing and validating a chemistry corpus from the Semantic Scholar Open Research Corpus using conservative, metadata-based license screening. Using this workflow, we assembled an internal study corpus of 582,683 chemistry-specific full-text research articles with structured full text, token-aware paragraph chunks, paragraph-level embeddings generated with the intfloat/e5-large-v2 model, and record-level metadata including abstracts and licensing information. To support downstream retrieval and text-mining use cases, an eligible subset of the corpus was additionally enriched with machine-generated brief summaries and multi-label subfield annotations spanning 18 chemistry domains. Licensing was screened using metadata from Unpaywall, OpenAlex, and Crossref, and the resulting corpus was technically validated for schema compliance, embedding reproducibility, text quality, and metadata completeness. The primary contribution of this work is a reproducible workflow for corpus construction and validation, together with its associated schema and reproducibility resources. The released materials include the code, reconstruction workflow, schema, metadata/provenance artifacts, and validation outputs needed to reproduce the corpus from pinned public upstream resources. Public redistribution of source-derived text and broad text-derived representations is outside the scope of the general release. Researchers can reproduce the workflow by using the released pipeline with publicly available upstream datasets and metadata services.

cs.DB↗

arXiv2026

Denoising and Baseline Correction of Low-Scan FTIR Spectra: A Benchmark of Deep Learning Models Against Traditional Signal Processing

High-quality Fourier Transform Infrared (FTIR) imaging usually needs extensive signal averaging to reduce noise and drift which severely limits clinical speed. Deep learning can accelerate imaging by reconstructing spectra from rapid, single-scan inputs. However, separating noise and baseline drift simultaneously without ground truth is an ill-posed inverse problem. Standard black-box architectures often rely on statistical approximations that introduce spectral hallucinations or fail to generalize to unstable atmospheric conditions. To solve these issues we propose a physics-informed cascade Unet that separates denoising and baseline correction tasks using a new, deterministic Physics Bridge. This architecture forces the network to separate random noise from chemical signals using an embedded SNIP layer to enforce spectroscopic constraints instead of learning statistical approximations. We benchmarked this approach against a standard single Unet and a traditional Savitzky-Golay/SNIP workflow. We used a dataset of human hypopharyngeal carcinoma cells (FaDu). The cascade model outperformed all other methods, achieving a 51.3% reduction in RMSE compared to raw single-scan inputs, surpassing both the single Unet (40.2%) and the traditional workflow (33.7%). Peak-aware metrics show that the cascade architecture eliminates spectral hallucinations found in standard deep learning. It also preserves peak intensity with much higher fidelity than traditional smoothing. These results show that the cascade Unet is a robust solution for diagnostic-grade FTIR imaging. It enables imaging speeds 32 times faster than current methods.

eess.IV↗

arXiv2025

Systematic Evaluation of Preprocessing Techniques for Accurate Image Registration in Digital Pathology

Image registration refers to the process of spatially aligning two or more images by mapping them into a common coordinate system, so that corresponding anatomical or tissue structures are matched across images. In digital pathology, registration enables direct comparison and integration of information from different stains or imaging modalities, sup-porting applications such as biomarker analysis and tissue reconstruction. Accurate registration of images from different modalities is an essential step in digital pathology. In this study, we investigated how various color transformation techniques affect image registration between hematoxylin and eosin (H&E) stained images and non-linear multimodal images. We used a dataset of 20 tissue sample pairs, with each pair undergoing several preprocessing steps, including different color transformation (CycleGAN, Macenko, Reinhard, Vahadane), inversion, contrast adjustment, intensity normalization, and denoising. All images were registered using the VALIS registration method, which first applies rigid registration and then performs non-rigid registration in two steps on both low and high-resolution images. Registration performance was evaluated using the relative Target Registration Error (rTRE). We reported the median of median rTRE values (MMrTRE) and the average of median rTRE values (AMrTRE) for each method. In addition, we performed a custom point-based evaluation using ten manually selected key points. Registration was done separately for two scenarios, using either the original or inverted multimodal images. In both scenarios, CycleGAN color transformation achieved the lowest registration errors, while the other methods showed higher errors. These findings show that applying color transformation before registration improves alignment between images from different modalities and supports more reliable analysis in digital pathology.

cs.CV↗

arXiv2025

Lightweight CycleGAN Models for Cross-Modality Image Transformation and Experimental Quality Assessment in Fluorescence Microscopy

Lightweight deep learning models offer substantial reductions in computational cost and environmental impact, making them crucial for scientific applications. We present a lightweight CycleGAN for modality transfer in fluorescence microscopy (confocal to super-resolution STED/deconvolved STED), addressing the common challenge of unpaired datasets. By replacing the traditional channel-doubling strategy in the U-Net-based generator with a fixed channel approach, we drastically reduce trainable parameters from 41.8 million to approximately nine thousand, achieving superior performance with faster training and lower memory usage. We also introduce the GAN as a diagnostic tool for experimental and labeling quality. When trained on high-quality images, the GAN learns the characteristics of optimal imaging; deviations between its generated outputs and new experimental images can reveal issues such as photobleaching, artifacts, or inaccurate labeling. This establishes the model as a practical tool for validating experimental accuracy and image fidelity in microscopy workflows.

cs.CV↗

arXiv2025

Global-to-local image quality assessment in optical microscopy via fast and robust deep learning predictions

Optical microscopy is one of the most widely used techniques in research studies for life sciences and biomedicine. These applications require reliable experimental pipelines to extract valuable knowledge from the measured samples and must be supported by image quality assessment (IQA) to ensure correct processing and analysis of the image data. IQA methods are implemented with variable complexity. However, while most quality metrics have a straightforward implementation, they might be time consuming and computationally expensive when evaluating a large dataset. In addition, quality metrics are often designed for well-defined image features and may be unstable for images out of the ideal domain. To overcome these limitations, recent works have proposed deep learning-based IQA methods, which can provide superior performance, increased generalizability and fast prediction. Our method, named $\mathrm{\mu}$DeepIQA, is inspired by previous studies and applies a deep convolutional neural network designed for IQA on natural images to optical microscopy measurements. We retrained the same architecture to predict individual quality metrics and global quality scores for optical microscopy data. The resulting models provide fast and stable predictions of image quality by generalizing quality estimation even outside the ideal range of standard methods. In addition, $\mathrm{\mu}$DeepIQA provides patch-wise prediction of image quality and can be used to visualize spatially varying quality in a single image. Our study demonstrates that optical microscopy-based studies can benefit from the generalizability of deep learning models due to their stable performance in the presence of outliers, the ability to assess small image patches, and rapid predictions.

cs.CV↗

arXiv2025

GFSR-Net: Guided Focus via Segment-Wise Relevance Network for Interpretable Deep Learning in Medical Imaging

Deep learning has achieved remarkable success in medical image analysis, however its adoption in clinical practice is limited by a lack of interpretability. These models often make correct predictions without explaining their reasoning. They may also rely on image regions unrelated to the disease or visual cues, such as annotations, that are not present in real-world conditions. This can reduce trust and increase the risk of misleading diagnoses. We introduce the Guided Focus via Segment-Wise Relevance Network (GFSR-Net), an approach designed to improve interpretability and reliability in medical imaging. GFSR-Net uses a small number of human annotations to approximate where a person would focus within an image intuitively, without requiring precise boundaries or exhaustive markings, making the process fast and practical. During training, the model learns to align its focus with these areas, progressively emphasizing features that carry diagnostic meaning. This guidance works across different types of natural and medical images, including chest X-rays, retinal scans, and dermatological images. Our experiments demonstrate that GFSR achieves comparable or superior accuracy while producing saliency maps that better reflect human expectations. This reduces the reliance on irrelevant patterns and increases confidence in automated diagnostic tools.

eess.IV↗

arXiv2025

LEO: An Open-Source Platform for Linking OMERO with Lab Notebooks and Heterogeneous Metadata Sources

In the interdisciplinary field of microscopy research, managing and integrating large volumes of data stored across disparate platforms remains a major challenge. Data types such as bioimages, experimental records, and spectral information are often maintained in separate repositories, each following different management standards. However, linking these data sources across the research lifecycle is essential to align with the FAIR principles of data management: Findability, Accessibility, Interoperability, and Reusability. Despite this need, there is a notable lack of tools capable of effectively integrating and linking data from heterogeneous sources. To address this gap, we present LEO (Linking Electronic Lab Notebooks with OMERO), a web-based platform designed to create and manage links between distributed data systems. LEO was initially developed to link objects between Electronic Lab Notebooks (ELNs) and OMERO, but its functionality has since been extended through a plugin-based architecture, allowing the integration of additional data sources. This extensibility makes LEO a scalable and flexible solution for a wide range of microscopy research workflows.

cs.CE↗

arXiv2025

Chunk Twice, Embed Once: A Systematic Study of Segmentation and Representation Trade-offs in Chemistry-Aware Retrieval-Augmented Generation

Retrieval-Augmented Generation (RAG) systems are increasingly vital for navigating the ever-expanding body of scientific literature, particularly in high-stakes domains such as chemistry. Despite the promise of RAG, foundational design choices -- such as how documents are segmented and represented -- remain underexplored in domain-specific contexts. This study presents the first large-scale, systematic evaluation of chunking strategies and embedding models tailored to chemistry-focused RAG systems. We investigate 25 chunking configurations across five method families and evaluate 48 embedding models on three chemistry-specific benchmarks, including the newly introduced QuestChemRetrieval dataset. Our results reveal that recursive token-based chunking (specifically R100-0) consistently outperforms other approaches, offering strong performance with minimal resource overhead. We also find that retrieval-optimized embeddings -- such as Nomic and Intfloat E5 variants -- substantially outperform domain-specialized models like SciBERT. By releasing our datasets, evaluation framework, and empirical benchmarks, we provide actionable guidelines for building effective and efficient chemistry-aware RAG systems.

cs.IR↗

arXiv2025

ChemQuests: A Curated Chemistry Question-Answer Database Extracted from ChemRxiv papers

The rapid expansion of chemistry literature poses significant challenges for researchers seeking to efficiently access domain-specific knowledge. To support advancements in chemistry-focused natural language processing (NLP), we present ChemQuests, a curated dataset of 952 high-quality question-answer (QA) pairs derived from 155 ChemRxiv \cite{chemrxivWebsite} papers across 17 subfields of chemistry. Each QA pair is explicitly linked to its source text segment to ensure traceability and contextual accuracy. ChemQuests was constructed using an automated pipeline that combines optical character recognition (OCR), QA generation using GPT-4o, and fuzzy-search verification. The dataset emphasizes conceptual, mechanistic, applied, and synthetic or experimental questions, enabling applications in retrieval-based QA systems, search engine development, and fine-tuning of domain-adapted large language models. We analyze the dataset's structure, coverage, and limitations, and outline future directions for expansion and expert validation. ChemQuests provides a foundational resource for chemistry NLP research, education, and tool development.

cs.AI↗

arXiv2025

Multi-Marker Similarity enables reduced-reference and interpretable image quality assessment in optical microscopy

Optical microscopy contributes to the ever-increasing progress in biological and biomedical studies, as it allows the implementation of minimally invasive experimental pipelines to translate the data of measured samples into valuable knowledge. Within these pipelines, reliable quality assessment must be ensured to validate the generated results. Image quality assessment is often applied with full-reference methods to estimate the similarity between the ground truth and the output images. However, current methods often show poor agreement with visual perception and lead to the generation of various full-reference metrics tailored to specific applications. Additionally, they rely on pixel-wise comparisons, emphasizing local intensity similarity while often overlooking comprehensive and interpretable image quality assessment. To address these issues, we have developed a multi-marker similarity method that compares standard quality markers, such as resolution, signal-to-noise ratio, contrast, and high frequency components. The method computes a similarity score between the image and the ground truth for each marker, then combines these scores into an overall similarity estimate. This provides a full-reference estimate of image quality while extracting global quality features and detecting experimental artifacts. Multi-marker similarity provides a reliable and interpretable method for image quality assessment and the generation of quality rankings. By focusing on the comparison of quality markers rather than direct image distances, the method enables reduced reference implementations, where a single field of view is used as a benchmark for multiple measurements. This opens the way for reliable automatic evaluation of big datasets, typical of large biomedical studies, when manual assessment of single images and defining the ground truth for each field of view is not feasible.

q-bio.QM↗

arXiv2025

Nonlinear Inference Capacity of Fiber-Optical Extreme Learning Machines

The intrinsic complexity of nonlinear optical phenomena offers a fundamentally new resource to analog brain-inspired computing, with the potential to address the pressing energy requirements of artificial intelligence. We introduce and investigate the concept of nonlinear inference capacity in optical neuromorphic computing in highly nonlinear fiber-based optical Extreme Learning Machines. We demonstrate that this capacity scales with nonlinearity to the point where it surpasses the performance of a deep neural network model with five hidden layers on a scalable nonlinear classification benchmark. By comparing normal and anomalous dispersion fibers under various operating conditions and against digital classifiers, we observe a direct correlation between the system's nonlinear dynamics and its classification performance. Our findings suggest that image recognition tasks, such as MNIST, are incomplete in showcasing deep computing capabilities in analog hardware. Our approach provides a framework for evaluating and comparing computational capabilities, particularly their ability to emulate deep networks, across different physical and digital platforms, paving the way for a more generalized set of benchmarks for unconventional, physics-inspired computing architectures.

physics.optics↗

arXiv2024

Investigating the effect of non-resonant background variation on the CARS data analysis and classification

: Non-resonant background (NRB) plays a significant role in coherent anti-Stokes Raman scattering (CARS) spectroscopic applications. All the recent works primarily focused on removing the NRB using different deep learning methods, and only one study explored the effect of NRB. Hence, in this work, we systematically investigated the impact of NRB variation on Raman signal retrieval. The NRB is simulated as a linear function with different strengths relative to the resonant Raman signal, and the variance also changed for each NRB strength. The resonant part of nonlinear susceptibility is extracted from real experimental Raman data; hence, the simulated CARS data better approximate the experimental CARS spectra. Then, the corresponding Raman signal is retrieved by four different methods: maximum entropy method (MEM), Kramers-Kronig (KK), convolutional neural network (CNN), and long short-term memory (LSTM) network. Pearson correlation measurements and principal component analysis combined with linear discriminant analysis (PCA-LDA) modelling revealed that MEM and KK methods have an edge over LSTM and CNN for higher NRB strengths. It is also demonstrated that normalizing the input data favors LSTM and CNN predictions. In contrast, background removal from the predictions significantly influenced Pearson correlation but not the classification accuracies for MEM and KK. This comprehensive study is done for the first time to the best of our knowledge and has the potential to impact the CARS spectroscopy and microscopy applications in different areas.

physics.optics↗

arXiv2022

RAMANMETRIX: a delightful way to analyze Raman spectra

Although Raman spectroscopy is widely used for the investigation of biomedical samples and has a high potential for use in clinical applications, it is not common in clinical routines. One of the factors that obstruct the integration of Raman spectroscopic tools into clinical routines is the complexity of the data processing workflow. Software tools that simplify spectroscopic data handling may facilitate such integration by familiarizing clinical experts with the advantages of Raman spectroscopy. Here, RAMANMETRIX is introduced as a user-friendly software with an intuitive web-based graphical user interface (GUI) that incorporates a complete workflow for chemometric analysis of Raman spectra, from raw data pretreatment to a robust validation of machine learning models. The software can be used both for model training and for the application of the pretrained models onto new data sets. Users have full control of the parameters during model training, but the testing data flow is frozen and does not require additional user input. RAMANMETRIX is available in two versions: as standalone software and web application. Due to the modern software architecture, the computational backend part can be executed separately from the GUI and accessed through an application programming interface (API) for applying a preconstructed model to the measured data. This opens up possibilities for using the software as a data processing backend for the measurement devices in real-time. The models preconstructed by more experienced users can be exported and reused for easy one-click data preprocessing and prediction, which requires minimal interaction between the user and the software. The results of such prediction and graphical outputs of the different data processing steps can be exported and saved.

physics.data-an↗

arXiv2015

Sample Size Planning for Classification Models

In biospectroscopy, suitably annotated and statistically independent samples (e. g. patients, batches, etc.) for classifier training and testing are scarce and costly. Learning curves show the model performance as function of the training sample size and can help to determine the sample size needed to train good classifiers. However, building a good model is actually not enough: the performance must also be proven. We discuss learning curves for typical small sample size situations with 5 - 25 independent samples per class. Although the classification models achieve acceptable performance, the learning curve can be completely masked by the random testing uncertainty due to the equally limited test sample size. In consequence, we determine test sample sizes necessary to achieve reasonable precision in the validation and find that 75 - 100 samples will usually be needed to test a good but not perfect classifier. Such a data set will then allow refined sample size planning on the basis of the achieved performance. We also demonstrate how to calculate necessary sample sizes in order to show the superiority of one classifier over another: this often requires hundreds of statistically independent test samples or is even theoretically impossible. We demonstrate our findings with a data set of ca. 2550 Raman spectra of single cells (five classes: erythrocytes, leukocytes and three tumour cell lines BT-20, MCF-7 and OCI-AML3) as well as by an extensive simulation that allows precise determination of the actual performance of the models in question.

stat.AP↗