[Explainable artificial intelligence in pathology].

With the advancements in precision medicine, the demands on pathological diagnostics have increased, requiring standardized, quantitative, and integrated assessments of histomorphological and molecular pathological data. Great hopes are placed in artificial intelligence (AI) methods, which have demonstrated the ability to analyze complex clinical, histological, and molecular data for disease classification, biomarker quantification, and prognosis estimation. This paper provides an overview of the latest developments in pathology AI, discusses the limitations, particularly concerning the black box character of AI, and describes solutions to make decision processes more transparent using methods of so-called explainable AI (XAI).

Prognostic value of deep learning-derived body composition in advanced pancreatic cancer-a retrospective multicenter study.

Background: Despite the prognostic relevance of cachexia in pancreatic cancer, individual body composition has not been routinely integrated into treatment planning. In this multicenter study, we investigated the prognostic value of sarcopenia and myosteatosis automatically extracted from routine computed tomography (CT) scans of patients with advanced pancreatic ductal adenocarcinoma (PDAC).

Patients And Methods: We retrospectively analyzed clinical imaging data of 601 patients from three German cancer centers.

Toward Explainable Artificial Intelligence for Precision Pathology.

The rapid development of precision medicine in recent years has started to challenge diagnostic pathology with respect to its ability to analyze histological images and increasingly large molecular profiling data in a quantitative, integrative, and standardized way. Artificial intelligence (AI) and, more precisely, deep learning technologies have recently demonstrated the potential to facilitate complex data analysis tasks, including clinical, histological, and molecular data for disease classification; tissue biomarker quantification; and clinical outcome prediction. This review provides a general introduction to AI and describes recent developments with a focus on applications in diagnostic pathology and beyond.

Single-cell gene regulatory network prediction by explainable AI.

The molecular heterogeneity of cancer cells contributes to the often partial response to targeted therapies and relapse of disease due to the escape of resistant cell populations. While single-cell sequencing has started to improve our understanding of this heterogeneity, it offers a mostly descriptive view on cellular types and states. To obtain more functional insights, we propose scGeneRAI, an explainable deep learning approach that uses layer-wise relevance propagation (LRP) to infer gene regulatory networks from static single-cell RNA sequencing data for individual cells.

DNA methylation-based classification of sinonasal tumors.

The diagnosis of sinonasal tumors is challenging due to a heterogeneous spectrum of various differential diagnoses as well as poorly defined, disputed entities such as sinonasal undifferentiated carcinomas (SNUCs). In this study, we apply a machine learning algorithm based on DNA methylation patterns to classify sinonasal tumors with clinical-grade reliability. We further show that sinonasal tumors with SNUC morphology are not as undifferentiated as their current terminology suggests but rather reassigned to four distinct molecular classes defined by epigenetic, mutational and proteomic profiles.

Patient-level proteomic network prediction by explainable artificial intelligence.

Understanding the pathological properties of dysregulated protein networks in individual patients' tumors is the basis for precision therapy. Functional experiments are commonly used, but cover only parts of the oncogenic signaling networks, whereas methods that reconstruct networks from omics data usually only predict average network features across tumors. Here, we show that the explainable AI method layer-wise relevance propagation (LRP) can infer protein interaction networks for individual patients from proteomic profiling data.

Differences in Characteristics of Error-Related Potentials Between Individuals With Spinal Cord Injury and Age- and Sex-Matched Able-Bodied Controls.

Non-invasive brain-computer interfaces (BCI) represent an emerging technology for enabling persons with impaired or lost grasping and reaching functions due to high spinal cord injury (SCI) to control assistive devices. A major drawback of BCIs is a high rate of false classifications. The robustness and performance of BCIs might be improved using cerebral electrophysiological correlates of error recognition (error-related potentials, ErrPs).

Sensory Feedback Interferes with Mu Rhythm Based Detection of Motor Commands from Electroencephalographic Signals.

Electroencephalogram (EEG)-based brain-computer interfaces (BCI) represent a promising component of restorative motor therapies in individuals with partial paralysis. However, in those patients, sensory functions such as proprioception are at least partly preserved. The aim of this study was to investigate whether afferent feedback interferes with the BCI-based detection of efferent motor commands during execution of movements.

Thrombolytic removal of intraventricular haemorrhage in treatment of severe stroke: results of the randomised, multicentre, multiregion, placebo-controlled CLEAR III trial.

Background: Intraventricular haemorrhage is a subtype of intracerebral haemorrhage, with 50% mortality and serious disability for survivors. We aimed to test whether attempting to remove intraventricular haemorrhage with alteplase versus saline irrigation improved functional outcome.

Methods: In this randomised, double-blinded, placebo-controlled, multiregional trial (CLEAR III), participants with a routinely placed extraventricular drain, in the intensive care unit with stable, non-traumatic intracerebral haemorrhage volume less than 30 mL, intraventricular haemorrhage obstructing the 3rd or 4th ventricles, and no underlying pathology were adaptively randomly assigned (1:1), via a web-based system to receive up to 12 doses, 8 h apart of 1 mg of alteplase or 0·9% saline via the extraventricular drain.

Safety and efficacy of minimally invasive surgery plus alteplase in intracerebral haemorrhage evacuation (MISTIE): a randomised, controlled, open-label, phase 2 trial.

Background: Craniotomy, according to the results from trials, does not improve functional outcome after intracerebral haemorrhage. Whether minimally invasive catheter evacuation followed by thrombolysis for clot removal is safe and can achieve a good functional outcome is not known. We investigated the safety and efficacy of alteplase, a recombinant tissue plasminogen activator, in combination with minimally invasive surgery (MIS) in patients with intracerebral haemorrhage.

A multicenter, randomized, double-blinded, placebo-controlled phase III study of Clot Lysis Evaluation of Accelerated Resolution of Intraventricular Hemorrhage (CLEAR III).

Background: In adults, intraventricular thrombolytic therapy with recombinant tissue plasminogen activator (rtPA) facilitates resolution of intraventricular haemorrhage (IVH), reduces intracranial pressure, decreases duration of cerebrospinal fluid diversion, and may ameliorate direct neural injury. We hypothesize that patients with small parenchymal haematoma volumes (<30 cc) and relatively large IVH causing acute obstructive hydrocephalus would have improved clinical outcomes when given injections of low-dose rtPA to accelerate lysis and evacuation of IVH compared with placebo.

Methods: The Clot Lysis Evaluation of Accelerated Resolution of Intraventricular Hemorrhage III trial is an investigator-initiated, phase III, randomized, multicenter, double-blind, placebo-controlled study comparing the use of external ventricular drainage (EVD) combined with intraventricular injection of rtPA to EVD plus intraventricular injection of normal saline (placebo) for the treatment of IVH.

Factors affecting clot lysis rates in patients with spontaneous intraventricular hemorrhage.

Background And Purpose: In the treatment of severe intraventricular hemorrhage (IVH), thrombolytic use and clot size are known to influence clot lysis rates. We evaluated the effect of other variables on IVH clot lysis rates among patients treated with recombinant tissue-type plasminogen activator or placebo.

Methods: One hundred patients with IVH and intracerebral hemorrhage volume <30 mL requiring emergency external ventricular drainage from 2 multicenter trials were treated with intraventricular administration of recombinant tissue-type plasminogen activator (n=78; 53 males, 25 females) or placebo (n=22; 7 males, 15 females).

Low-dose recombinant tissue-type plasminogen activator enhances clot resolution in brain hemorrhage: the intraventricular hemorrhage thrombolysis trial.

Background And Purpose: Patients with intracerebral hemorrhage and intraventricular hemorrhage have a reported mortality of 50% to 80%. We evaluated a clot lytic treatment strategy for these patients in terms of mortality, ventricular infection, and bleeding safety events, and for its effect on the rate of intraventricular clot lysis.

Methods: Forty-eight patients were enrolled at 14 centers and randomized to treatment with 3 mg recombinant tissue-type plasminogen activator (rtPA) or placebo.

Preliminary report of the clot lysis evaluating accelerated resolution of intraventricular hemorrhage (CLEAR-IVH) clinical trial.

Introduction: Brain hemorrhage is the most frequent fatal form of stroke and has the highest level of morbidity of any stroke subtype. For patients with both intracerebral hemorrhage and intraventricular hemorrhage (IVH), expected mortality is 50-80%. No validated, efficacious treatment exists for humans, but animal models demonstrate substantial physiologic and functional benefits associated with rapid, near-complete removal of blood from either the ventricle or intracerebral location (i.

Preliminary findings of the minimally-invasive surgery plus rtPA for intracerebral hemorrhage evacuation (MISTIE) clinical trial.

Introduction: Compared to ischemic stroke, intracerebral hemorrhage (ICH) is easily and rapidly identified, occurs in younger patients, and produces relatively small initial injury to cerebral tissues--all factors suggesting that interventional amelioration is possible. Investigations from the last decade established that extent of ICH-mediated brain injury relates directly to blood clot volume and duration of blood exposure to brain tissue. Using minimally-invasive surgery plus recombinant tissue plasminogen activator (rtPA), MISTIE investigators explored aggressive avenues to treat ICH.

Visual fields at follow-up in the Ischemic Optic Neuropathy Decompression Trial: evaluation of change in pattern defect and severity over time.

Purpose: To evaluate change from baseline to 12 months follow-up in study and nonstudy (fellow) eye visual fields from the Ischemic Optic Neuropathy Decompression Trial (IONDT).

Design: Randomized controlled trial and observational study.

Participants: The IONDT enrolled patients >or=50 years with acute nonarteritic ischemic optic neuropathy (NAION).

A prospective, population-based study of the role of visual impairment in motor vehicle crashes among older drivers: the SEE study.

Purpose: To determine the role of vision and visual attention factors in automobile crash involvement.

Methods: Drivers aged 65 to 84 years were identified during the baseline interview (1993-1995) of the Salisbury Eye Evaluation (SEE) Study. Crash involvement through December 1997 was determined from Maryland State motor vehicle records.

Disparities in the emergency department evaluation of chest pain patients.

Background: The existence of race and gender differences in the provision of cardiovascular health care has been increasingly recognized. However, few studies have examined whether these differences exist in the emergency department (ED) setting.

Objectives: To evaluate race, gender, and insurance differences in the receipt of early, noninvasive diagnostic tests among persons presenting to an ED with a complaint of chest pain.

Frameless stereotactic aspiration and thrombolysis of spontaneous intracerebral hemorrhage.

Introduction: To test the feasibility and safety of a minimally invasive technique, we report our experience in treating spontaneous intracerebral hemorrhage (ICH) patients by using frameless stereotactic clot aspiration-thrombolysis and its effects on their 30-day survival. We compared the observed cohort mortality with its predicted 30-day ICH mortality, by using previously validated methods.

Methods: Selection criteria were diagnosis of hypertensive ICH > or =35 cc, reduced level of consciousness, and no brainstem compression.

Target-specific catecholamine elevation induced by anticonvulsant thalamic deep brain stimulation.

Purpose: Anterior thalamic nucleus (AN) deep brain stimulation (DBS) is effective in raising EEG and clonic seizure threshold in experimental models. Little is known about the specific properties of DBS that afford its anticonvulsant effect. We sought to test the hypothesis that experimental seizures and the anticonvulsant action of AN DBS alter the underlying regional neurochemistry of AN, specifically with facilitation of the serotonergic system to local electrical stimulation.

Intraventricular thrombolysis speeds blood clot resolution: results of a pilot, prospective, randomized, double-blind, controlled trial.

Objective: Animal models and clinical studies suggest that intraventricular thrombolysis improves clot resolution and clinical outcomes among patients with intraventricular hemorrhage. However, this intervention may increase the rates of rebleeding and infection. To assess the safety and efficacy of intraventricular thrombolysis, we conducted a pilot, randomized, double-blind, controlled, multicenter study.

The role of physicians in assessing older drivers: barriers, opportunities, and strategies.

Background: Evaluation of the older driver is a difficult task for primary care physicians. We investigated the physician-perceived barriers to assessing older drivers in primary care practice.

Methods: Twenty family physicians whose patients had completed a clinical questionnaire and neuropsychological tests participated in one of 2 focus groups.

A longitudinal study of risk factors for motorcycle crashes among junior college students in Taiwan.

A cohort of 4729 junior college students in an urban and a rural area in Taiwan was followed up for a period of 20 months. Students' characteristics, including riding exposures, as well as human, vehicular, and environmental factors were collected using one initial and three follow-up questionnaires. The Anderson-Gill (AG) multiplicative intensity model was used to determine the risk of a motorcycle crash over time while also allowing for the modeling of multiple events.

Prospective study of brief neuropsychological measures to assess crash risk in older primary care patients.

Background: Practicing primary care physicians often encounter the difficult clinical situation of evaluating the older driver. We wanted to investigate the relation between self-reported driving behavior, neuropsychological measures, and crash risk to inform the development of a test battery that could predict unsafe driving behavior and was feasible for use by primary care physicians.

Methods: This study was a prospective follow-up of 107 drivers aged 65 years and older recruited from a primary care setting in 1995.