Computed Tomography (CT) to Cone-Beam Computed Tomography (CBCT) image registration is crucial for image-guided radiotherapy and surgical procedures. However, achieving accurate CT-CBCT registration remains challenging due to various factors such as inconsistent intensities, low contrast resolution and imaging artifacts. In this study, we propose a Context-Aware Semantics-driven Hierarchical Network (referred to as CASHNet), which hierarchically integrates context-aware semantics-encoded features into a coarse-to-fine registration scheme, to explicitly enhance semantic structural perception during progressive alignment. Moreover, it leverages diffeomorphisms to integrate rigid and non-rigid registration within a single end-to-end trainable network, enabling anatomically plausible deformations and preserving topological consistency. CASHNet comprises a Siamese Mamba-based multi-scale feature encoder and a coarse-to-fine registration decoder, which integrates a Rigid Registration (RR) module with multiple Semantics-guided Velocity Estimation and Feature Alignment (SVEFA) modules operating at different resolutions. Each SVEFA module comprises three carefully designed components: i) a cross-resolution feature aggregation (CFA) component that synthesizes enhanced global contextual representations, ii) a semantics perception and encoding (SPE) component that captures and encodes local semantic information, and iii) an incremental velocity estimation and feature alignment (IVEFA) component that leverages contextual and semantic features to update velocity fields and to align features. These modules work synergistically to boost the overall registration performance. Extensive experiments on three typical yet challenging CT-CBCT datasets of both soft and hard tissues demonstrate the superiority of our proposed method over other state-of-the-art methods. The code will be publicly available at https://github.com/xiaorugao999/CASHNet.

Universal medical image registration through a single model handling various registration tasks has attracted increasing interest. However, existing deep learning-based methods face two major challenges in adapting to universal registration tasks: 1) they lack generalizable feature representation capabilities for cross-task registration; 2) they rely solely on model architectures with fixed parameters, which limits their flexibility to dynamically adapt to different registration tasks and inherently compromises their generalization capability for zero-shot performance on unseen tasks. To address these limitations, we propose CIM-VTP, a novel two-stage universal registration framework. In the first stage, our proposed Correlation-guided Image Modeling (CIM)-based pretraining strategy leverages cross-image correlation to guide the masked modeling process, which facilitates spatial correspondence capturing that is essential for registration and provides universal representation capabilities as a foundation for registration learning. In the second stage, we introduce a registration task classifier to identify the type of a given input task, which explicitly quantifies the similarity between current inputs and previously seen tasks. The obtained task similarity scores are then fed as prior information into our carefully designed multi-resolution Visual-Textual Task Prompt (VTP) modules, which integrate task-relevant knowledge through prompt learning to adaptively adjust decoder parameters for different input domains. Extensive experiments across six different registration tasks demonstrate that the proposed CIM-VTP exhibits superior universal image registration performance. The code will be released at https://github.com/xiehousheng/CIM-VTP.

The field of computer vision applied to videos of minimally invasive surgery is ever-growing. Workflow recognition pertains to the automated recognition of various aspects of a surgery, including: which surgical steps are performed; and which surgical instruments are used. This information can later be used to assist clinicians when learning the surgery or during live surgery. The Pituitary Vision (PitVis) 2023 Challenge tasks the community to step and instrument recognition in videos of endoscopic pituitary surgery. This is a particularly challenging task when compared to other minimally invasive surgeries due to: the smaller working space, which limits and distorts vision; and higher frequency of instrument and step switching, which requires more precise model predictions. Participants were provided with 25-videos, with results presented at the MICCAI-2023 conference as part of the Endoscopic Vision 2023 Challenge in Vancouver, Canada, on 08-Oct-2023. There were 18-submissions from 9-teams across 6-countries, using a variety of deep learning models. The top performing model for step recognition utilised a transformer based architecture, uniquely using an autoregressive decoder with a positional encoding input. The top performing model for instrument recognition utilised a spatial encoder followed by a temporal encoder, which uniquely used a 2-layer temporal architecture. In both cases, these models outperformed purely spatial based models, illustrating the importance of sequential and temporal information. This PitVis-2023 therefore demonstrates state-of-the-art computer vision models in minimally invasive surgery are transferable to a new dataset. Benchmark results are provided in the paper, and the dataset is publicly available at: https://doi.org/10.5522/04/26531686.

In a clinical setting, the acquisition of certain medical image modality is often unavailable due to various considerations such as cost, radiation, etc. Therefore, unpaired cross-modality translation techniques, which involve training on the unpaired data and synthesizing the target modality with the guidance of the acquired source modality, are of great interest. Previous methods for synthesizing target medical images are to establish one-shot mapping through generative adversarial networks (GANs). As promising alternatives to GANs, diffusion models have recently received wide interests in generative tasks. In this paper, we propose a target-guided diffusion model (TGDM) for unpaired cross-modality medical image translation. For training, to encourage our diffusion model to learn more visual concepts, we adopted a perception prioritized weight scheme (P2W) to the training objectives. For sampling, a pre-trained classifier is adopted in the reverse process to relieve modality-specific remnants from source data. Experiments on both brain MRI-CT and prostate MRI-US datasets demonstrate that the proposed method achieves a visually realistic result that mimics a vivid anatomical section of the target organ. In addition, we have also conducted a subjective assessment based on the synthesized samples to further validate the clinical value of TGDM.

When developing context-aware systems, automatic surgical phase recognition and tool presence detection are two essential tasks. There exist previous attempts to develop methods for both tasks but majority of the existing methods utilize a frame-level loss function (e.g., cross-entropy) which does not fully leverage the underlying semantic structure of a surgery, leading to sub-optimal results. In this paper, we propose multi-task learning-based, LAtent Space-constrained Transformers, referred as LAST, for automatic surgical phase recognition and tool presence detection. Our design features a two-branch transformer architecture with a novel and generic way to leverage video-level semantic information during network training. This is done by learning a non-linear compact presentation of the underlying semantic structure information of surgical videos through a transformer variational autoencoder (VAE) and by encouraging models to follow the learned statistical distributions. In other words, LAST is of structure-aware and favors predictions that lie on the extracted low dimensional data manifold. Validated on two public datasets of the cholecystectomy surgery, i.e., the Cholec80 dataset and the M2cai16 dataset, our method achieves better results than other state-of-the-art methods. Specifically, on the Cholec80 dataset, our method achieves an average accuracy of 93.12±4.71%, an average precision of 89.25±5.49%, an average recall of 90.10±5.45% and an average Jaccard of 81.11 ±7.62% for phase recognition, and an average mAP of 95.15±3.87% for tool presence detection. Similar superior performance is also observed when LAST is applied to the M2cai16 dataset.

In manyclinical settings, a lot of medical image datasets suffer from imbalance problems, which makes predictions of trained models to be biased toward majority classes. Semi-supervised Learning (SSL) algorithms trained with such imbalanced datasets become more problematic since pseudo-supervision of unlabeled data are generated from the model's biased predictions. To address these issues, in this work, we propose a novel semi-supervised deep learning method, i.e., uncertainty-guided virtual adversarial training (VAT) with batch nuclear-norm (BNN) optimization, for large-scale medical image classification. To effectively exploit useful information from both labeled and unlabeled data, we leverage VAT and BNN optimization to harness the underlying knowledge, which helps to improve discriminability, diversity and generalization of the trained models. More concretely, our network is trained by minimizing a combination of four types of losses, including a supervised cross-entropy loss, a BNN loss defined on the output matrix of labeled data batch (lBNN loss), a negative BNN loss defined on the output matrix of unlabeled data batch (uBNN loss), and a VAT loss on both labeled and unlabeled data. We additionally propose to use uncertainty estimation to filter out unlabeled samples near the decision boundary when computing the VAT loss. We conduct comprehensive experiments to evaluate the performance of our method on two publicly available datasets and one in-house collected dataset. The experimental results demonstrated that our method achieved better results than state-of-the-art SSL methods.

In this paper, we propose an end-to-end deep learning architecture, referred as MCG-Net, integrating convolutional neural network (CNN) with transformer-based global context block for fine-grained delineation and diagnostic classification of four cardiac events from magnetocardiogram (MCG) data, namely Q-, R-, S- and T-waves. MCG-Net takes advantage of a multi-resolution CNN backbone as well as the state-of-the-art (SOTA) transformer encoders that facilitate global temporal feature aggregation. Besides the novel network architecture, we introduce a multi-task learning scheme to achieve simultaneous delineation and classification. Specifically, the problem of MCG delineation is formulated as multi-class heatmap regression. Meanwhile, a binary diagnostic classification label as well as a duration are jointly estimated for each cardiac event using features that are temporally aligned by event heatmaps. The framework is evaluated on a clinical MCG dataset, containing data collected from 270 subjects with cardiac anomalies and 108 control subjects. We designed and conducted a two-fold cross-validation study to validate the proposed method and to compare its performance with the SOTA methods. Experimental results demonstrated that our method outperformed counterparts on both event delineation and diagnostic classification tasks, achieving respectively an average ECG-F1 of 0.987 and an average Event-F1 of 0.975 for MCG delineation, and an average accuracy of 0.870, an average sensitivity of 0.732, an average specificity of 0.914 and an average AUC of 0.903 for diagnostic classification. Comprehensive ablation experiments are additionally performed to investigate effectiveness of different network components.

Many different methods have been proposed for generation of synthetic CT (sCT) from MR images. Most of these methods depend on paired-wise aligned MR and CT training images of the same patient, which are difficult to obtain. In this paper, we propose a novel disentangled representation learning method for MR to CT synthesis using unpaired data. Specifically, we first embed images onto two spaces: a modality-invariant geometry space capturing the shared anatomical information across different imaging domains, and a modality-specific appearance space. From the embedding, a sCT image can be synthesized from a MR image by taking the encoded geometry features from the MR image and an appearance vector sampled from the appearance space of a CT image. To handle the challenging of distinguishing cortical bone from air in MR images, where both of them have low intensity values, we propose a novel Geometry Similarity Module (GSM) to take the context information into consideration. Experimental results demonstrated that our approach achieved better or equivalent results than the state-of-the-art.

Nonlinear regression has been extensively employed in many computer vision problems (e.g., crowd counting, age estimation, affective computing). Under the umbrella of deep learning, two common solutions exist i) transforming nonlinear regression to a robust loss function which is jointly optimizable with the deep convolutional network, and ii) utilizing ensemble of deep networks. Although some improved performance is achieved, the former may be lacking due to the intrinsic limitation of choosing a single hypothesis and the latter may suffer from much larger computational complexity. To cope with those issues, we propose to regress via an efficient "divide and conquer" manner. The core of our approach is the generalization of negative correlation learning that has been shown, both theoretically and empirically, to work well for non-deep regression problems. Without extra parameters, the proposed method controls the bias-variance-covariance trade-off systematically and usually yields a deep regression ensemble where each base model is both "accurate" and "diversified." Moreover, we show that each sub-problem in the proposed method has less Rademacher Complexity and thus is easier to optimize. Extensive experiments on several diverse and challenging tasks including crowd counting, personality analysis, age estimation, and image super-resolution demonstrate the superiority over challenging baselines as well as the versatility of the proposed method. The source code and trained models are available on our project page: https://mmcheng.net/dncl/.

Reports on changes to the author information presented in the above named paper.

The papers in this special section focus on multi-modal computing in biomedical applications. n recent years, the development of biomedical imaging techniques, integrative sensors, and machine learning, brings many benefits to the diagnosis of various diseases. We can collect, measure, and analyze vast volumes of health-related data using the technologies of computing and networking, leading to tremendous opportunities for the health and biomedical community. Meanwhile, these technologies have also brought new challenges and issues. Biomedical intelligence, especially precision medicine, is considered one of the most promising directions for healthcare development.

Quantification of cerebral white matter hyperintensities (WMH) of presumed vascular origin is of key importance in many neurological research studies. Currently, measurements are often still obtained from manual segmentations on brain MR images, which is a laborious procedure. The automatic WMH segmentation methods exist, but a standardized comparison of the performance of such methods is lacking. We organized a scientific challenge, in which developers could evaluate their methods on a standardized multi-center/-scanner image dataset, giving an objective comparison: the WMH Segmentation Challenge. Sixty T1 + FLAIR images from three MR scanners were released with the manual WMH segmentations for training. A test set of 110 images from five MR scanners was used for evaluation. The segmentation methods had to be containerized and submitted to the challenge organizers. Five evaluation metrics were used to rank the methods: 1) Dice similarity coefficient; 2) modified Hausdorff distance (95th percentile); 3) absolute log-transformed volume difference; 4) sensitivity for detecting individual lesions; and 5) F1-score for individual lesions. In addition, the methods were ranked on their inter-scanner robustness; 20 participants submitted their methods for evaluation. This paper provides a detailed analysis of the results. In brief, there is a cluster of four methods that rank significantly better than the other methods, with one clear winner. The inter-scanner robustness ranking shows that not all the methods generalize to unseen scanners. The challenge remains open for future submissions and provides a public platform for method evaluation.

Accurate segmentation of infant brain magnetic resonance (MR) images into white matter (WM), gray matter (GM), and cerebrospinal fluid (CSF) is an indispensable foundation for early studying of brain growth patterns and morphological changes in neurodevelopmental disorders. Nevertheless, in the isointense phase (approximately 6-9 months of age), due to inherent myelination and maturation process, WM and GM exhibit similar levels of intensity in both T1-weighted (T1w) and T2-weighted (T2w) MR images, making tissue segmentation very challenging. Despite many efforts were devoted to brain segmentation, only few studies have focused on the segmentation of 6-month infant brain images. With the idea of boosting methodological development in the community, iSeg-2017 challenge (http://iseg2017.web.unc.edu) provides a set of 6-month infant subjects with manual labels for training and testing the participating methods. Among the 21 automatic segmentation methods participating in iSeg-2017, we review the 8 top-ranked teams, in terms of Dice ratio, modified Hausdorff distance and average surface distance, and introduce their pipelines, implementations, as well as source codes. We further discuss limitations and possible future directions. We hope the dataset in iSeg-2017 and this review article could provide insights into methodological development for the community.

Deep convolutional networks (ConvNets) have achieved unprecedented performances on many computer vision tasks. However, their adaptations to crowd counting on single images are still in their infancy and suffer from severe over-fitting. Here we propose a new learning strategy to produce generalizable features by way of deep negative correlation learning (NCL). More specifically, we deeply learn a pool of decorrelated regressors with sound generalization capabilities through managing their intrinsic diversities. Our proposed method, named decorrelated ConvNet (D-ConvNet), is end-to-end-trainable and independent of the backbone fully-convolutional network architectures. Extensive experiments on very deep VGGNet as well as our customized network structure indicate the superiority of D-ConvNet when compared with several state-of-the-art methods. Our implementation will be released at https://github.com/shizenglin/Deep-NCL.

Reconstruction of the anterior cruciate ligament (ACL) through arthroscopy is one of the most common procedures in orthopaedics. It requires accurate alignment and drilling of the tibial and femoral tunnels through which the ligament graft is attached. Although commercial computer-assisted navigation systems exist to guide the placement of these tunnels, most of them are limited to a fixed pose without due consideration of dynamic factors involved in different knee flexion angles. This paper presents a new model for intraoperative guidance of arthroscopic ACL reconstruction with reduced error particularly in the ligament attachment area. The method uses 3D preoperative data at different flexion angles to build a subject-specific statistical model of knee pose. To circumvent the problem of limited training samples and ensure physically meaningful pose instantiation, homogeneous transformations between different poses and local-deformation finite element modelling are used to enlarge the training set. Subsequently, an anatomical geodesic flexion analysis is performed to extract the subject-specific flexion characteristics. The advantages of the method were also tested by detailed comparison to standard Principal Component Analysis (PCA), nonlinear PCA without training set enlargement, and other state-of-the-art articulated joint modelling methods. The method yielded sub-millimetre accuracy, demonstrating its potential clinical value.

This paper addresses the problem of fully-automatic localization and segmentation of 3D intervertebral discs (IVDs) from MR images. Our method contains two steps, where we first localize the center of each IVD, and then segment IVDs by classifying image pixels around each disc center as foreground (disc) or background. The disc localization is done by estimating the image displacements from a set of randomly sampled 3D image patches to the disc center. The image displacements are estimated by jointly optimizing the training and test displacement values in a data-driven way, where we take into consideration both the training data and the geometric constraint on the test image. After the disc centers are localized, we segment the discs by classifying image pixels around disc centers as background or foreground. The classification is done in a similar data-driven approach as we used for localization, but in this segmentation case we are aiming to estimate the foreground/background probability of each pixel instead of the image displacements. In addition, an extra neighborhood smooth constraint is introduced to enforce the local smoothness of the label field. Our method is validated on 3D T2-weighted turbo spin echo MR images of 35 patients from two different studies. Experiments show that compared to state of the art, our method achieves better or comparable results. Specifically, we achieve for localization a mean error of 1.6-2.0 mm, and for segmentation a mean Dice metric of 85%-88% and a mean surface distance of 1.3-1.4 mm.

Cephalometric analysis is an essential clinical and research tool in orthodontics for the orthodontic analysis and treatment planning. This paper presents the evaluation of the methods submitted to the Automatic Cephalometric X-Ray Landmark Detection Challenge, held at the IEEE International Symposium on Biomedical Imaging 2014 with an on-site competition. The challenge was set to explore and compare automatic landmark detection methods in application to cephalometric X-ray images. Methods were evaluated on a common database including cephalograms of 300 patients aged six to 60 years, collected from the Dental Department, Tri-Service General Hospital, Taiwan, and manually marked anatomical landmarks as the ground truth data, generated by two experienced medical doctors. Quantitative evaluation was performed to compare the results of a representative selection of current methods submitted to the challenge. Experimental results show that three methods are able to achieve detection rates greater than 80% using the 4 mm precision range, but only one method achieves a detection rate greater than 70% using the 2 mm precision range, which is the acceptable precision range in clinical practice. The study provides insights into the performance of different landmark detection approaches under real-world conditions and highlights achievements and limitations of current image analysis techniques.

An accurate determination of the pelvic orientation is inevitable for the correct cup prosthesis placement of navigated total hip arthroplasties. Conventionally, this step is accomplished by percutaneous palpation of anatomic landmarks. Sterility issues and an increased landmark localization error for obese patients lead to the application of B-mode ultrasound imaging in the field of computer-assisted orthopedic surgery. Many approaches have been proposed in the literature to replace the percutaneous digitization by 3-D B-mode ultrasound imaging. However, the correct depth localization of the pelvic landmarks could be significantly affected by the acoustic properties of the penetrated tissues. Imprecise depth estimation could lead to a miscalculation of the pelvic orientation and subsequently to a misalignment of the acetabular cup implant. But so far, no solution has been presented, which compensates for acoustic property differences for correct depth estimation. In this paper, we present a novel approach to determine pelvic orientation from ultrasound images by applying a hierarchical registration scheme based on patch statistical shape models to compensate for differences in speed of sound. The method was validated based on plastic bones and a cadaveric specimen.

International audience

In this paper, we present a novel robotic assistant dedicated to otologic surgery for an implantable hearing aid system, which is a procedure involving drilling into the lateral skull of the patient. This compact and flexible miniature robot is designed so as to fulfill the requirements of precise bone drillings for hearing aid implantation. It is built from an original five degree-of-freedom (DOF) parallel structure with a motorized end-effector, particularly well suited to otological surgical procedure. The specification, the design and the analysis of the workspace are detailed. A preliminary accuracy evaluation is presented. A rotational accuracy of 0.6 ± 0.6° and a translational accuracy of 1.4 ± 0.5 mm were found.

In this paper we generalize the belief propagation based rigid shape matching algorithm to a nonparametric belief propagation based on parameterized shape matching. We construct a local-global shape descriptor based cost function to compare the distances among landmarks in each data set, which is equivalent to the Hamiltonian of a spin glass. The constructed cost function is immune to rigid transformations, therefore the parameterized shape matching can be achieved by searching for the optimal shape parameter and the correspondence assignment that minimize the cost function. The optimization procedure is then approximated by a Monte Carlo simulation based MAP estimation on a graphical model, i.e. the nonparametric belief propagation. Experiments on a principal component analysis (PCA) based point distribution model (PDM) of the proximal femur illustrate the effects of two key factors, the topology of the graphical model and the renormalization of the shape parameters of the parameterized shape. Other factors that can influence its performance and its computational complexity are also discussed.

The widely used procedure of evaluation of cup orientation following THA using single standard anteroposterior radiograph is known inaccurate, largely due to the wide variability in individual pelvic position relative to X-ray plate. 3D-2D image registration methods have been introduced to estimate the transformation between a CT volume and the radiograph for an accurate estimation of the cup orientation relative to an anatomical reference extracted from the CT data. However, the robustness of these methods is questionable. This paper presents a robust image similarity measure which is derived from a variational approximation to mutual information and allows for incorporation of spatial information via energy minimization. Experimental results on estimating cup alignment from single X-ray radiograph with gonadal shielding demonstrate the robustness and the accuracy of the present approach.

It has been recognized that one of the most difficult steps in intramedullary nailing of femoral shaft fractures is the distal locking - the insertion of distal transverse interlocking screws, for which it is necessary to know the positions and orientations of the distal locking holes (DLHs) of the intramedullary nail (IMN). This paper presents a robust and accurate approach for solving this problem based on two calibrated and registered fluoroscopic images. The problem is formulated as a two-stage model-based optimal fitting process. The first stage, nail detection, automatically estimates the axis of the distal part of the IMN (DP-IMN) by iteratively fitting a cylindrical model to the images. The second stage, pose recovery, resolves the translations and the rotations of the DLHs around the estimated axis by iteratively fitting the geometrical models of the DLHs to the images. An iterative best matched projection point (IBMPP) algorithm is combined with random sample strategies to effectively and robustly solve the fitting problems in both stages. We designed and conducted comprehensive experiments to validate the robustness and the accuracy of the present approach. Our in vitro experiments show on average less than 14 s execution time on a Linux machine, a mean angular error of 0.48 degrees (std = 0.21 degrees ), and a mean translational error of 0.09 mm (std = 0.041 mm). We conclude that the present approach is fast, robust, and accurate for distal locking applications.

Automatic identification and extraction of bone contours from x-ray images is an essential first step task for further medical image analysis. In this paper we propose a 3D statistical model based framework for the proximal femur contour extraction from calibrated x-ray images. The automatic initialization is solved by an Estimation of Bayesian Network Algorithm to fit a multiple component geometrical model to the x-ray data. The contour extraction is accomplished by a non-rigid 2D/3D registration between a 3D statistical model and the x-ray images, in which bone contours are extracted by a graphical model based Bayesian inference. Preliminary experiments on clinical data sets verified its validity.

Surface reconstruction of bone from a few X-ray images and point distribution model (PDM) is discussed. We present a robust approach combining regularized morphing and shape deformation, and show its application to surface reconstruction of proximal femur. The robustness of the presented approach relies on the development of a novel method to establish correspondence between the X-ray images and the model estimated from the PDM. The correspondence is based on an iterative non-rigid twodimensional (2D) point matching process, which iteratively uses a symmetric injective nearest-neighbor mapping operator (SIN-MO) and 2D thin-plate splines (2D-TPS) based deformation to find a fraction of best matched 2D point pairs between edge points detected from the X-ray images and projections of the points on the apparent contours extracted from the estimated model. The advantages of this novel method include robustness with respect to outliers, and automatic exclusion of cross matching, which is an important property for preservation of topology. Initial quantitative and qualitative evaluation results are given which indicate the validity of our approach.

We address the problem of automatic reconstruction of patient-specific 3D surface model of proximal femur from biplanar calibrated fluoroscopic images. Previously, we proposed a point distribution model (PDM) based reconstruction algorithm which incorporates an original 2D/3D correspondence building method to convert a 2D/3D surface reconstruction problem to a 3D/3D one. The convergence of this algorithm relies on a proper initialization. In our previous work, interactively reconstructed anatomical landmarks were used for this purpose. In contrast, this paper presents a fully automatic initialization method, which uses a particle filter based inference algorithm to automatically determine the geometrical parameters of a proximal femur from its biplanar fluoroscopic images. The estimated geometrical parameters are then used to initialize the reconstruction algorithm. Here we report the quantitative and qualitative evaluation results on 10 dry cadaveric bones. Compared to the manual initialization, the automated initialization results in a little bit less accurate reconstruction but has the advantage of elimination of user interactions

Reconstruction of patient-specific three-dimensional (3D) bone model from biplanar two-dimensional (2D) X-ray images and point distribution models (PDM) is discussed. We present a stable and accurate approach combining regularized morphing and shape deformation, and show its application to reconstruction of proximal femur. A novel image-to-model correspondence building method using directly the edge pixels detected from the 2D images and the apparent contour extracted from the 3D model is proposed to convert a 2D/3D reconstruction problem to a 3D/3D one, whose solutions are well studied. Quantitative and qualitative evaluation results on eleven cadaveric dry bones are given which indicate the validity of our approach.