How do AI markings on screening mammograms correspond to cancer location? An informed review of 270 breast cancer cases in BreastScreen Norway.

OBJECTIVES: To compare the location of AI markings on screening mammograms with cancer location on diagnostic mammograms, and to classify interval cancers with high AI score as false negative, minimal sign, or true negative. METHODS: In a retrospective study from 2022, we compared the performance of an AI system with independent double reading according to cancer detection. We found 93% (880/949) of the screen-detected cancers, and 40% (122/305) of the interval cancers to have the highest AI risk score (AI score of 10). In this study, four breast radiologists reviewed mammograms from 126 randomly selected screen-detected cancers and all 120 interval cancers with an AI score of 10. The location of the AI marking was stated as correct/not correct in craniocaudal and mediolateral oblique view. Interval cancers with an AI score of 10 were classified as false negative, minimal sign significant/non-specific, or true negative. RESULTS: All screen-detected cancers and 78% (93/120) of the interval cancers with an AI score of 10 were correctly located by the AI system. The AI markings matched in both views for 79% (100/126) of the screen-detected cancers and 22% (26/120) of the interval cancers. For interval cancers with an AI score of 10, 11% (13/120) were correctly located and classified as false negative, 10% (12/120) as minimal sign significant, 26% (31/120) as minimal sign non-specific, and 31% (37/120) as true negative. CONCLUSION: AI markings corresponded to cancer location for all screen-detected cancers and 78% of the interval cancers with high AI score, indicating a potential for reducing the number of interval cancers. However, it is uncertain whether interval cancers with subtle findings in only one view are actionable for recall in a true screening setting. CLINICAL RELEVANCE STATEMENT: In this study, AI markings corresponded to the location of the cancer in a high percentage of cases, indicating that the AI system accurately identifies the cancer location in mammograms with a high AI score. KEY POINTS: • All screen-detected and 78% of the interval cancers with high AI risk score (AI score of 10) had AI markings in one or two views corresponding to the location of the cancer on diagnostic images. • Among all 120 interval cancers with an AI score of 10, 21% (25/120) were classified as a false negative or minimal sign significant and had AI markings matching the cancer location, suggesting they may be visible on prior screening. • Most of the correctly located interval cancers matched only in one view, and the majority were classified as either true negative or minimal sign non-specific, indicating low potential for being detected earlier in a real screening setting.

AI performance by mammographic density in a retrospective cohort study of 99,489 participants in BreastScreen Norway.

OBJECTIVE: To explore the ability of artificial intelligence (AI) to classify breast cancer by mammographic density in an organized screening program. MATERIALS AND METHOD: We included information about 99,489 examinations from 74,941 women who participated in BreastScreen Norway, 2013-2019. All examinations were analyzed with an AI system that assigned a malignancy risk score (AI score) from 1 (lowest) to 10 (highest) for each examination. Mammographic density was classified into Volpara density grade (VDG), VDG1-4; VDG1 indicated fatty and VDG4 extremely dense breasts. Screen-detected and interval cancers with an AI score of 1-10 were stratified by VDG. RESULTS: We found 10,406 (10.5% of the total) examinations to have an AI risk score of 10, of which 6.7% (704/10,406) was breast cancer. The cancers represented 89.7% (617/688) of the screen-detected and 44.6% (87/195) of the interval cancers. 20.3% (20,178/99,489) of the examinations were classified as VDG1 and 6.1% (6047/99,489) as VDG4. For screen-detected cancers, 84.0% (68/81, 95% CI, 74.1-91.2) had an AI score of 10 for VDG1, 88.9% (328/369, 95% CI, 85.2-91.9) for VDG2, 92.5% (185/200, 95% CI, 87.9-95.7) for VDG3, and 94.7% (36/38, 95% CI, 82.3-99.4) for VDG4. For interval cancers, the percentages with an AI score of 10 were 33.3% (3/9, 95% CI, 7.5-70.1) for VDG1 and 48.0% (12/25, 95% CI, 27.8-68.7) for VDG4. CONCLUSION: The tested AI system performed well according to cancer detection across all density categories, especially for extremely dense breasts. The highest proportion of screen-detected cancers with an AI score of 10 was observed for women classified as VDG4. CLINICAL RELEVANCE STATEMENT: Our study demonstrates that AI can correctly classify the majority of screen-detected and about half of the interval breast cancers, regardless of breast density. KEY POINTS: • Mammographic density is important to consider in the evaluation of artificial intelligence in mammographic screening. • Given a threshold representing about 10% of those with the highest malignancy risk score by an AI system, we found an increasing percentage of cancers with increasing mammographic density. • Artificial intelligence risk score and mammographic density combined may help triage examinations to reduce workload for radiologists.

Core-ionization spectrum of liquid water.

We present state-of-the-art calculations of the core-ionization spectrum of water. Despite significant progress in procedures developed to mitigate various experimental complications and uncertainties, the experimental determination of ionization energies of solvated species involves several non-trivial steps such as assessing the effect of the surface potential, electrolytes, and finite escape depths of photoelectrons. This provides a motivation to obtain robust theoretical values of the intrinsic bulk ionization energy and the corresponding solvent-induced shift. Here we develop theoretical protocols based on coupled-cluster theory and electrostatic embedding. Our value of the intrinsic solvent-induced shift of the 1sO ionization energy of water is -1.79 eV. The computed absolute position and the width of the 1sO peak in photoelectron spectrum of water are 538.47 eV and 1.44 eV, respectively, agreeing well with the best experimental values.

Prediction of photodynamics of 200 nm excited cyclobutanone with linear response electronic structure and ab initio multiple spawning.

Simulations of photochemical reaction dynamics have been a challenge to the theoretical chemistry community for some time. In an effort to determine the predictive character of current approaches, we predict the results of an upcoming ultrafast diffraction experiment on the photodynamics of cyclobutanone after excitation to the lowest lying Rydberg state (S2). A picosecond of nonadiabatic dynamics is described with ab initio multiple spawning. We use both time dependent density functional theory (TDDFT) and equation-of-motion coupled cluster singles and doubles (EOM-CCSD) theory for the underlying electronic structure theory. We find that the lifetime of the S2 state is more than a picosecond (with both TDDFT and EOM-CCSD). The predicted ultrafast electron diffraction spectrum exhibits numerous structural features, but weak time dependence over the course of the simulations.

Time-Dependent Multilevel Density Functional Theory.

We present a novel three-layer approach based on multilevel density functional theory (MLDFT) and polarizable molecular mechanics to simulate the electronic excitations of chemical systems embedded in an external environment within the time-dependent DFT formalism. In our method, the electronic structure of a target system, the chromophore, is determined in the field of an embedded inactive layer, which is treated as frozen. Long-range interactions are described by employing the polarizable fluctuating charge (FQ) force field. The resulting MLDFT/FQ thus accurately describes both electrostatics (and polarization) and non-electrostatic target-environment interactions. The robustness and reliability of the approach are demonstrated by comparing our results with experimental data reported for various organic molecules in solution.

Coupled Cluster Theory for Nonadiabatic Dynamics: Nuclear Gradients and Nonadiabatic Couplings in Similarity Constrained Coupled Cluster Theory.

Coupled cluster theory is one of the most accurate electronic structure methods for predicting ground and excited state chemistry. However, the presence of numerical artifacts at electronic degeneracies, such as complex energies, has made it difficult to apply the method in nonadiabatic dynamics simulations. While it has already been shown that such numerical artifacts can be fully removed by using similarity constrained coupled cluster (SCC) theory [J. Phys. Chem. Lett. 2017, 8(19), 4801-4807], simulating dynamics requires efficient implementations of gradients and nonadiabatic couplings. Here, we present an implementation of nuclear gradients and nonadiabatic derivative couplings at the similarity constrained coupled cluster singles and doubles (SCCSD) level of theory, thereby making possible nonadiabatic dynamics simulations using a coupled cluster theory that provides a correct description of conical intersections between excited states. We present a few numerical examples that show good agreement with literature values and discuss some limitations of the method.

Collective Strong Coupling Modifies Aggregation and Solvation.

Intermolecular (Coulombic) interactions are pivotal for aggregation, solvation, and crystallization. We demonstrate that the collective strong coupling of several molecules to a single optical mode results in notable changes in the molecular excitations around a single perturbed molecule, thus representing an impurity in an otherwise ordered system. A competition between short-range coulombic and long-range photonic correlations inverts the local transition density in a polaritonic state, suggesting notable changes in the polarizability of the solvation shell. Our results provide an alternative perspective on recent work in polaritonic chemistry and pave the way for the rigorous treatment of cooperative effects in aggregation, solvation, and crystallization.

Understanding X-ray absorption in liquid water using triple excitations in multilevel coupled cluster theory.

X-ray absorption (XA) spectroscopy is an essential experimental tool to investigate the local structure of liquid water. Interpretation of the experiment poses a significant challenge and requires a quantitative theoretical description. High-quality theoretical XA spectra require reliable molecular dynamics simulations and accurate electronic structure calculations. Here, we present the first successful application of coupled cluster theory to model the XA spectrum of liquid water. We overcome the computational limitations on system size by employing a multilevel coupled cluster framework for large molecular systems. Excellent agreement with the experimental spectrum is achieved by including triple excitations in the wave function and using molecular structures from state-of-the-art path-integral molecular dynamics. We demonstrate that an accurate description of the electronic structure within the first solvation shell is sufficient to successfully model the XA spectrum of liquid water within the multilevel framework. Furthermore, we present a rigorous charge transfer analysis of the XA spectrum, which is reliable due to the accuracy and robustness of the electronic structure methodology. This analysis aligns with previous studies regarding the character of the prominent features of the XA spectrum of liquid water.

Quantum Mechanical Versus Polarizable Embedding Schemes: A Study of the Xray Absorption Spectra of Aqueous Ammonia and Ammonium.

The X-ray absorption spectra of aqueous ammonia and ammonium are computed using a combination of coupled cluster singles and doubles (CCSD) with different quantum mechanical and molecular mechanical embedding schemes. Specifically, we compare frozen Hartree-Fock (HF) density embedding, polarizable embedding (PE), and polarizable density embedding (PDE). Integrating CCSD with frozen HF density embedding is possible within the CC-in-HF framework, which circumvents the conventional system-size limitations of standard coupled cluster methods. We reveal similarities between PDE and frozen HF density descriptions, while PE spectra differ significantly. By including approximate triple excitations, we also investigate the effect of improving the electronic structure theory. The spectra computed using this approach show an improved intensity ratio compared to CCSD-in-HF. Charge transfer analysis of the excitations shows the local character of the pre-edge and main-edge, while the post-edge is formed by excitations delocalized over the first solvation shell and beyond.