Catalytic enantioconvergent coupling of secondary and tertiary electrophiles with olefins.

Carbon-carbon bonds, including those between sp3-hybridized carbon atoms (alkyl-alkyl bonds), typically comprise much of the framework of organic molecules. In the case of sp3-hybridized carbon, the carbon can be stereogenic and the particular stereochemistry can have implications for structure and function1-3. As a consequence, the development of methods that simultaneously construct alkyl-alkyl bonds and control stereochemistry is important, although challenging. Here we describe a strategy for enantioselective alkyl-alkyl bond formation, in which a racemic alkyl electrophile is coupled with an olefin in the presence of a hydrosilane, rather than via a traditional electrophile-nucleophile cross-coupling, through the action of a chiral nickel catalyst. We demonstrate that families of racemic alkyl halides-including secondary and tertiary electrophiles, which have not previously been shown to be suitable for enantioconvergent coupling with alkyl metal nucleophiles-cross-couple with olefins with good enantioselectivity and yield under very mild reaction conditions. Given the ready availability of olefins, our approach opens the door to developing more general methods for enantioconvergent alkyl-alkyl coupling.

Association of preoperative frailty with the risk of postoperative delirium in older patients undergoing hip fracture surgery: a prospective cohort study.

OBJECTIVE: This study aimed to explore the correlation between preoperative frailty and the risk of postoperative delirium (POD) in older patients undergoing hip fracture surgery. METHODS: In total, 148 patients with hip fractures who were admitted to Tsinghua Changgung Hospital (Beijing, China) between January 2022 and January 2023 were involved in this study. Preoperative frailty scales were assessed, of which the CAM scale was postoperatively administered every morning and evening on days 1, 2, 3, 5, and 7. Binary logistic regression analysis was conducted to determine the correlation between preoperative frailty and the risk of POD. RESULTS: Among 148 older patients with hip fractures, 71 (48.0%) were identified as preoperative frail and 77 (52.0%) as non-frail. The overall incidence of POD on day 7 was 24.3% (36/148), and preoperative frailty was associated with a significantly higher risk of POD compared with non-frailty (42.3% vs. 7.8%, P < 0.001). The binary logistic regression analysis revealed that preoperative frailty was noted as an independent risk factor for the risk of POD in older patients undergoing hip fracture surgery (P = 0.002). CONCLUSION: Preoperative frailty increased the risk of POD in older patients undergoing hip fracture surgery. DISCUSSION: Preoperative assessment of frailty in geriatric hip surgery can timely identify potential risks and provide interventions targeting frailty factors to reduce the incidence of POD in older patients undergoing hip fracture surgery. The findings suggested that preoperative frailty could increase the risk of POD in older patients undergoing hip fracture surgery. Further research is necessary to determine whether perioperative interventions aimed at enhancing frailty can mitigate the risk of POD and improve prognosis in older patients undergoing hip fracture surgery.

Position Checking-Based Sampling Approach Combined with Attraction Point Local Optimization for Safe Flight of UAVs.

Trading off the allocation of limited computational resources between front-end path generation and back-end trajectory optimization plays a key role in improving the efficiency of unmanned aerial vehicle (UAV) motion planning. In this paper, a sampling-based kinodynamic planning method that can reduce the computational cost as well as the risks of UAV flight is proposed. Firstly, an initial trajectory connecting the start and end points without considering obstacles is generated. Then, a spherical space is constructed around the topological vertices of the environment, based on the intersections of the trajectory with the obstacles. Next, some unnecessary sampling points, as well as node rewiring, are discarded by the designed position-checking strategy to minimize the computational cost and reduce the risks of UAV flight. Finally, in order to make the planning framework adaptable to complex scenarios, the strategies for selecting different attraction points according to the environment are designed, which further ensures the safe flight of the UAV while improving the success rate of the front-end trajectory. Simulations and real-world experiment comparisons are conducted on a vision-based platform to verify the performance of the proposed method.

Investigation of the C-N Bond-Forming Step in a Photoinduced, Copper-Catalyzed Enantioconvergent N-Alkylation: Characterization and Application of a Stabilized Organic Radical as a Mechanistic Probe.

Whereas photoinduced, copper-catalyzed couplings of nitrogen nucleophiles with alkyl electrophiles have recently been shown to provide an attractive approach to achieving a variety of enantioselective C-N bond constructions, mechanistic studies of these transformations have lagged the advances in reaction development. Herein we provide mechanistic insight into a previously reported photoinduced, copper-catalyzed enantioconvergent C-N coupling of a carbazole nucleophile with a racemic tertiary α-haloamide electrophile. Building on the isolation of a copper(II) model complex whose EPR parameters serve as a guide, we independently synthesize two key intermediates in the proposed catalytic cycle, a copper(II) metalloradical (L*CuII(carb')2) (L* = a monodentate chiral phosphine ligand; carb' = a carbazolide ligand), as well as a tertiary α-amide organic radical (R·); the generation and characterization of R· was guided by DFT calculations, which suggested that it would be stable to homocoupling. Continuous-wave (CW) and pulse EPR studies, along with corresponding DFT calculations, are among the techniques used to characterize these reactive radicals. We establish that these two radicals do indeed combine to furnish the C-N coupling product in good yield and with significant enantiomeric excess (77% yield, 55% ee), thereby supporting the chemical competence of these proposed intermediates. DFT calculations are consistent with R· initially binding to copper(II) via a dative interaction from the closed-shell carbonyl oxygen atom of the radical, which positions the α-carbon for direct reaction with the copper(II)-bound carbazole N atom, to generate the C-N bond with enantioselectivity, without the formation of an alkylcopper(III) intermediate.

A Magnetic Resonance Angiography-Based Study Comparing Machine Learning and Clinical Evaluation: Screening Intracranial Regions Associated with the Hemorrhagic Stroke of Adult Moyamoya Disease.

OBJECTIVES: Moyamoya disease patients with hemorrhagic stroke usually have a poor prognosis. This study aimed to determine whether hemorrhagic moyamoya disease could be distinguished from MRA images using transfer deep learning and to screen potential regions that contain rich distinguishing information from MRA images in moyamoya disease. MATERIALS AND METHODS: A total of 116 adult patients with bilateral moyamoya diseases suffering from hemorrhagic or ischemia complications were retrospectively screened. Based on original MRA images at the level of the basal cistern, basal ganglia, and centrum semiovale, we adopted the pretrained ResNet18 to build three models for differentiating hemorrhagic moyamoya disease. Grad-CAM was applied to visualize the regions of interest. RESULTS: For the test set, the accuracies of model differentiation in the basal cistern, basal ganglia, and centrum semiovale were 93.3%, 91.5%, and 86.4%, respectively. Visualization of the regions of interest demonstrated that the models focused on the deep and periventricular white matter and abnormal collateral vessels in hemorrhagic moyamoya disease. CONCLUSION: A transfer learning model based on MRA images of the basal cistern and basal ganglia showed a good ability to differentiate between patients with hemorrhagic moyamoya disease and those with ischemic moyamoya disease. The deep and periventricular white matter and collateral vessels at the level of the basal cistern and basal ganglia may contain rich distinguishing information.

Synergistic N-Heterocyclic Carbene/Palladium-Catalyzed Umpolung 1,4-Addition of Aryl Iodides to Enals.

An umpolung 1,4-addition of aryl iodides to enals promoted by cooperative (terpy)Pd/NHC catalysis was developed that generates various bioactive ß,ß-diaryl propanoate derivatives. This system is not only the first reported palladium-catalyzed arylation of NHC-bound homoenolates but also expands the scope of NHC-induced umpolung transformations. A diverse array of functional groups such as esters, nitriles, alcohols, and heterocycles are tolerated under the mild conditions. This method also circumvents the use of moisture-sensitive organometallic reagents.

Mechanistic Investigation of Enantioconvergent Kumada Reactions of Racemic α-Bromoketones Catalyzed by a Nickel/Bis(oxazoline) Complex.

In recent years, a wide array of methods for achieving nickel-catalyzed substitution reactions of alkyl electrophiles by organometallic nucleophiles, including enantioconvergent processes, have been described; however, experiment-focused mechanistic studies of such couplings have been comparatively scarce. The most detailed mechanistic investigations to date have examined catalysts that bear tridentate ligands and, with one exception, processes that are not enantioselective; studies of catalysts based on bidentate ligands could be anticipated to be more challenging, due to difficulty in isolating proposed intermediates as a result of instability arising from coordinative unsaturation. In this investigation, we explore the mechanism of enantioconvergent Kumada reactions of racemic α-bromoketones catalyzed by a nickel complex that bears a bidentate chiral bis(oxazoline) ligand. Utilizing an array of mechanistic tools (including isolation and reactivity studies of three of the four proposed nickel-containing intermediates, as well as interrogation via EPR spectroscopy, UV-vis spectroscopy, radical probes, and DFT calculations), we provide support for a pathway in which carbon-carbon bond formation proceeds via a radical-chain process wherein a nickel(I) complex serves as the chain-carrying radical and an organonickel(II) complex is the predominant resting state of the catalyst. Computations indicate that the coupling of this organonickel(II) complex with an organic radical is the stereochemistry-determining step of the reaction.

Understanding and Controlling the Emission Brightness and Color of Molecular Cerium Luminophores.

Molecular cerium complexes are a new class of tunable and energy-efficient visible- and UV-luminophores. Understanding and controlling the emission brightness and color are important for tailoring them for new and specialized applications. Herein, we describe the experimental and computational analyses for series of tris(guanidinate) (1-8, Ce{(R2N)C(N iPr)2}3, R = alkyl, silyl, or phenyl groups), guanidinate-amide [GA, A = N(SiMe3)2, G = (Me3Si)2NC(N iPr)2], and guanidinate-aryloxide (GOAr, OAr = 2,6-di- tert-butylphenoxide) cerium(III) complexes to understand and develop predictive capabilities for their optical properties. Structural studies performed on complexes 1-8 revealed marked differences in the steric encumbrance around the cerium center induced by various guanidinate ligand backbone substituents, a property that was correlated to photoluminescent quantum yield. Computational studies revealed that consecutive replacements of the amide and aryloxide ligands by guanidinate ligand led to less nonradiative relaxation of bright excited states and smaller Stokes shifts. The results establish a comprehensive structure-luminescence model for molecular cerium(III) luminophores in terms of both quantum yields and colors. The results provide a clear basis for the design of tunable, molecular, cerium-based, luminescent materials.

Cerium Photosensitizers: Structure-Function Relationships and Applications in Photocatalytic Aryl Coupling Reactions.

Two complete mixed-ligand series of luminescent Ce(III) complexes with the general formulas [(Me3Si)2NC(N(i)Pr)2]xCe(III)[N(SiMe3)2]3-x (x = 0, 1-N; x = 1, 2-N, x = 2, 3-N; x = 3, 4) and [(Me3Si)2NC(N(i)Pr)2]xCe(III)(OAr)3-x (x = 0, 1-OAr; x = 1, 2-OAr, x = 2, 3-OAr; x = 3, 4) were developed, featuring photoluminescence quantum yields up to 0.81(2) and lifetimes to 117(1) ns. Although the 4f → 5d absorptive transitions for these complexes were all found at ca. 420 nm, their emission bands exhibited large Stokes shifts with maxima occurring at 553 nm for 1-N, 518 nm for 2-N, 508 nm for 3-N, and 459 nm for 4, featuring yellow, lime-green, green, and blue light, respectively. Combined time-dependent density functional theory (TD-DFT) calculations and spectroscopic studies suggested that the long-lived (2)D excited states of these complexes corresponded to singly occupied 5dz(2) orbitals. The observed difference in the Stokes shifts was attributed to the relaxation of excited states through vibrational processes facilitated by the ligands. The photochemistry of the sterically congested complex 4 was demonstrated by C-C bond forming reaction between 4-fluoroiodobenzene and benzene through an outer sphere electron transfer pathway, which expands the capabilities of cerium photosensitizers beyond our previous results that demonstrated inner sphere halogen atom abstraction reactivity by 1-N.

The Hexachlorocerate(III) Anion: A Potent, Benchtop Stable, and Readily Available Ultraviolet A Photosensitizer for Aryl Chlorides.

The hexachlorocerate(III) anion, [CeIIICl6]3-, was found to be a potent photoreductant in acetonitrile solution with an estimated excited-state reduction potential of -3.45 V versus Cp2Fe0/+. Despite a short lifetime of 22.1(1) ns, the anion exhibited a photoluminescence quantum yield of 0.61(4) and fast quenching kinetics toward organohalogens allowing for its application in the photocatalytic reduction of aryl chloride substrates.

Cerium(III) and Uranium(IV) Complexes of the 2-Fluorophenyl Trimethylsilyl Amide Ligand: C-F → Ln/An Interactions that Modulate the Coordination Spheres of f-Block Elements.

2-fluorophenyl trimethylsilyl amide, N(SiMe3)(C6H4F)(-) was shown to engage in stronger C-F → Ce(III) interactions than pentafluorophenyl trimethylsilyl amide, N(SiMe3)(C6F5)(-), through a comparative study of the Ce(III) model complexes Ce[N(SiMe3)(C6H4F)]3 (1-F1) and Ce[N(SiMe3)(C6F5)]3 (1-F5). The presence of multiple C-F → U(IV) interactions led to complexes 2-X (X = Cl, C≡CPh, OMe) with threefold geometries, featuring a trigonal pyramidal UN3Cl core in the solid-state structures. Density functional theory calculations were applied to 2-Cl to investigate the strength of the C-F → U(IV) interactions and the influence of such interactions on resulting geometries.

Luminescent Ce(III) Complexes as Stoichiometric and Catalytic Photoreductants for Halogen Atom Abstraction Reactions.

Luminescent Ce(III) complexes, Ce[N(SiMe3)2]3 (1) and [(Me3Si)2NC(RN)2]Ce[N(SiMe3)2]2 (R = (i)Pr, 1-(i)Pr; R = Cy, 1-Cy), with C(3v) and C(2v) solution symmetries display absorptive 4f → 5d electronic transitions in the visible region. Emission bands are observed at 553, 518, and 523 nm for 1, 1-(i)Pr, and 1-Cy with lifetimes of 24, 67, and 61 ns, respectively. Time-dependent density functional theory (TD-DFT) studies on 1 and 1-(i)Pr revealed the (2)A1 excited states corresponded to singly occupied 5d(z(2)) orbitals. The strongly reducing metalloradical character of 1, 1-(i)Pr, and 1-Cy in their (2)A1 excited states afforded photochemical halogen atom abstraction reactions from sp(3) and sp(2) C-X (X = Cl, Br, I) bonds for the first time with a lanthanide cation. The dehalogenation reactions could be turned over with catalytic amounts of photosensitizers by coupling salt metathesis and reduction to the photopromoted atom abstraction reactions.

Investigation of Uranium Tris(imido) Complexes: Synthesis, Characterization, and Reduction Chemistry of [U(NDIPP)3(thf)3].

Addition of KC8 to trivalent [UI3(thf)4] in the presence of three equivalents of 2,6-diisopropylphenylazide (N3DIPP) results in the formation of the hexavalent uranium tris(imido) complex [U(NDIPP)3(thf)3] (1) through a facile, single-step synthesis. The X-ray crystal structure shows an octahedral complex that adopts a facial orientation of the imido substituents. This structural trend is maintained during the single-electron reduction of 1 to form dimeric [U(NDIPP)3{K(Et2O)}]2 (2). Variable-temperature/field magnetization studies of 2 show two independent U(V) 5f (1) centers, with no antiferromagnetic coupling present. Characterization of these complexes was accomplished using single-crystal X-ray diffraction, variable-temperature (1)H NMR spectroscopy, as well as IR and UV/Vis absorption spectroscopic studies.

NiXantphos: a deprotonatable ligand for room-temperature palladium-catalyzed cross-couplings of aryl chlorides.

Although the past 15 years have witnessed the development of sterically bulky and electron-rich alkylphosphine ligands for palladium-catalyzed cross-couplings with aryl chlorides, examples of palladium catalysts based on either triarylphosphine or bidentate phosphine ligands for efficient room temperature cross-coupling reactions with unactivated aryl chlorides are rare. Herein we report a palladium catalyst based on NiXantphos, a deprotonatable chelating aryldiphosphine ligand, to oxidatively add unactivated aryl chlorides at room temperature. Surprisingly, comparison of an extensive array of ligands revealed that under the basic reaction conditions the resultant heterobimetallic Pd-NiXantphos catalyst system outperformed all the other mono- and bidentate ligands in a deprotonative cross-coupling process (DCCP) with aryl chlorides. The DCCP with aryl chlorides affords a variety of triarylmethane products, a class of compounds with various applications and interesting biological activity. Additionally, the DCCP exhibits remarkable chemoselectivity in the presence of aryl chloride substrates bearing heteroaryl groups and sensitive functional groups that are known to undergo 1,2-addition, aldol reaction, and O-, N-, enolate-α-, and C(sp(2))-H arylations. The advantages and importance of the Pd-NiXantphos catalyst system outlined herein make it a valuable contribution for applications in Pd-catalyzed arylation reactions with aryl chlorides.

Anomalous one-electron processes in the chemistry of uranium nitrogen multiple bonds.

Novel reaction pathways are illustrated in the synthesis of uranium(IV), uranium(V), and uranium(VI) monoimido complexes. In contrast to the straightforward preparation of U(V)(═NSiMe3)[N(SiMe3)2]3 (1), the synthesis of a uranium(V) tritylimido complex, U(V)(═NCPh3)[N(SiMe3)2]3 (4), from U(III)[N(SiMe3)2]3 and Ph3CN3 was found to proceed through multiple one-electron steps. Whereas the oxidation of 1 with copper(II) salts produced the uranium(VI) monoimido complexes U(VI)(═NSiMe3)X[N(SiMe3)2]3 (X = Cl, Br), the reaction of 4 with CuBr2 undergoes sterically induced reduction to form the uranium(VI) monoimido complex U(VI)(═NCPh3)Br2[N(SiMe3)2]2, demonstrating a striking difference in reactivity based on imido substituent. The facile reduction of compounds 1 and 4 with KC8 allowed for the synthesis of the uranium(IV) monoimido derivatives, K[U(IV)(═NSiMe3)[N(SiMe3)2]3] (1-K) and K[U(IV)(═NCPh3)[N(SiMe3)2]3] (4-K), respectively. In contrast, an analogous uranium(IV) monoimido complex, K[U(IV)(═NPh(F))[N(SiMe3)Ph(F)]], Ph(F) = -pentafluorophenyl (6), was prepared through a loss of N(SiMe3)2Ph(F) concomitant with one-electron oxidation of a uranium(III) center. The uranium(IV) monoimido complexes were found to be reactive toward electrophiles, demonstrating N-C and N-Si single bond formation. One-electron reduction of nitrite provided a route to the uranium(VI) oxo/imido complex, [Ph4P][U(VI)O(═NSiMe3)[N(SiMe3)2]3]. The energetics and electrochemical processes involved in the various oxidation reactions are discussed. Finally, comparison of the U(VI)(═NSiMe3)X[N(SiMe3)2]3, X = Cl, Br, complexes with the previously reported U(VI)OX[N(SiMe3)2]3, X = Cl, Br, complexes suggested that the donor strength of the trimethylsilylimido ligand is comparable to the oxo ligand.

A new class of organocatalysts: sulfenate anions.

Sulfenate anions are known to act as highly reactive species in the organic arena. Now they premiere as organocatalysts. Proof of concept is offered by the sulfoxide/sulfenate-catalyzed (1-10 mol%) coupling of benzyl halides in the presence of base to generate trans-stilbenes in good to excellent yields (up to 99%). Mechanistic studies support the intermediacy of sulfenate anions, and the deprotonated sulfoxide was determined to be the resting state of the catalyst.

Electrophilic Ln(III) cations protected by C-F → Ln interactions and their coordination chemistry with weak σ- and π-donors.

A homoleptic cerium(III) amide complex, Ce(NPh(F)2)3 (1-Ce) (Ph(F) = pentafluorophenyl), in an unusual pseudo-trigonal planar geometry featuring six C-F → Ce interactions was prepared. The C-F → Ln interactions in solution were evident by comparison of the (19)F NMR shifts for the paramagnetic 1-Ce with those of the 4f(0) lanthanum(III) analogue. Coordination of weak σ- and π-donors, including ethers and neutral arene molecules, was achieved by the reversible displacement of the weak C-F → Ce interactions. Computational studies on Ce(NPh(F)2)3 and Ce(NPh(F)2)3(η(6)-C6H3Me3) provide information on the F → Ce interactions and Ce-η(6)-arene bonding.

Combined diagnosis of multiparametric MRI-based deep learning models facilitates differentiating triple-negative breast cancer from fibroadenoma magnetic resonance BI-RADS 4 lesions.

PURPOSE: To investigate the value of the combined diagnosis of multiparametric MRI-based deep learning models to differentiate triple-negative breast cancer (TNBC) from fibroadenoma magnetic resonance Breast Imaging-Reporting and Data System category 4 (BI-RADS 4) lesions and to evaluate whether the combined diagnosis of these models could improve the diagnostic performance of radiologists. METHODS: A total of 319 female patients with 319 pathologically confirmed BI-RADS 4 lesions were randomly divided into training, validation, and testing sets in this retrospective study. The three models were established based on contrast-enhanced T1-weighted imaging, diffusion-weighted imaging, and T2-weighted imaging using the training and validation sets. The artificial intelligence (AI) combination score was calculated according to the results of three models. The diagnostic performances of four radiologists with and without AI assistance were compared with the AI combination score on the testing set. The area under the curve (AUC), sensitivity, specificity, accuracy, and weighted kappa value were calculated to assess the performance. RESULTS: The AI combination score yielded an excellent performance (AUC = 0.944) on the testing set. With AI assistance, the AUC for the diagnosis of junior radiologist 1 (JR1) increased from 0.833 to 0.885, and that for JR2 increased from 0.823 to 0.876. The AUCs of senior radiologist 1 (SR1) and SR2 slightly increased from 0.901 and 0.950 to 0.925 and 0.975 after AI assistance, respectively. CONCLUSION: Combined diagnosis of multiparametric MRI-based deep learning models to differentiate TNBC from fibroadenoma magnetic resonance BI-RADS 4 lesions can achieve comparable performance to that of SRs and improve the diagnostic performance of JRs.

Noninvasive assessment of breast cancer molecular subtypes on multiparametric MRI using convolutional neural network with transfer learning.

BACKGROUND: To evaluate the performances of multiparametric MRI-based convolutional neural networks (CNNs) for the preoperative assessment of breast cancer molecular subtypes. METHODS: A total of 136 patients with 136 pathologically confirmed invasive breast cancers were randomly divided into training, validation, and testing sets in this retrospective study. The CNN models were established based on contrast-enhanced T1 -weighted imaging (T1 C), Apparent diffusion coefficient (ADC), and T2 -weighted imaging (T2 W) using the training and validation sets. The performances of CNN models were evaluated on the testing set. The area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and accuracy were calculated to assess the performance. RESULTS: For the separation of each subtype from other subtypes on the testing set, the T1 C-based models yielded AUCs from 0.762 to 0.920; the ADC-based models yielded AUCs from 0.686 to 0.851; and the T2 W-based models achieved AUCs from 0.639 to 0.697. CONCLUSION: T1 C-based models performed better than ADC-based models and T2 W-based models in assessing the breast cancer molecular subtypes. The discriminating performances of our CNN models for triple negative and human epidermal growth factor receptor 2-enriched subtypes were better than that of luminal A and luminal B subtypes.

Apparent Diffusion Coefficient-Based Convolutional Neural Network Model Can Be Better Than Sole Diffusion-Weighted Magnetic Resonance Imaging to Improve the Differentiation of Invasive Breast Cancer From Breast Ductal Carcinoma In Situ.

BACKGROUND AND PURPOSE: Breast ductal carcinoma in situ (DCIS) has no metastatic potential, and has better clinical outcomes compared with invasive breast cancer (IBC). Convolutional neural networks (CNNs) can adaptively extract features and may achieve higher efficiency in apparent diffusion coefficient (ADC)-based tumor invasion assessment. This study aimed to determine the feasibility of constructing an ADC-based CNN model to discriminate DCIS from IBC. METHODS: The study retrospectively enrolled 700 patients with primary breast cancer between March 2006 and June 2019 from our hospital, and randomly selected 560 patients as the training and validation sets (ratio of 3 to 1), and 140 patients as the internal test set. An independent external test set of 102 patients during July 2019 and May 2021 from a different scanner of our hospital was selected as the primary cohort using the same criteria. In each set, the status of tumor invasion was confirmed by pathologic examination. The CNN model was constructed to discriminate DCIS from IBC using the training and validation sets. The CNN model was evaluated using the internal and external tests, and compared with the discriminating performance using the mean ADC. The area under the curve (AUC), sensitivity, specificity, and accuracy were calculated to evaluate the performance of the previous model. RESULTS: The AUCs of the ADC-based CNN model using the internal and external test sets were larger than those of the mean ADC (AUC: 0.977 vs. 0.866, P = 0.001; and 0.926 vs. 0.845, P = 0.096, respectively). Regarding the internal test set and external test set, the ADC-based CNN model yielded sensitivities of 0.893 and 0.873, specificities of 0.929 and 0.894, and accuracies of 0.907 and 0.902, respectively. Regarding the two test sets, the mean ADC showed sensitivities of 0.845 and 0.818, specificities of 0.821 and 0.829, and accuracies of 0.836 and 0.824, respectively. Using the ADC-based CNN model, the prediction only takes approximately one second for a single lesion. CONCLUSION: The ADC-based CNN model can improve the differentiation of IBC from DCIS with higher accuracy and less time.